Berkeley Talks transcript: Professor Michael Omi on racial classification in the census

Intro: Hello OLLI. How’s everybody doing today? Good. All right, well, welcome to the Osher Lifelong Learning Institute’s, fourth and final winter 2019 speaker series event. Today’s speaker is Michael Omi and his talk is entitled Who Are You?: Racial Classification in the Census . Michael Omi is the co-author of Racial Formation in the United States , a groundbreaking work that transformed how we understand the social and historical forces that give race its changing meaning over time and place.

Since 1995, he’s been the co-editor of the book series on Asian American history and culture at Temple University Press. From 1999 to 2008, he served as a member and chair of the Daniel E. Koshland Committee for Civic Unity of the San Francisco Foundation. Since 2002, he served on the Project Advisory Board on Race and Human Variation for the American Anthropological Association that resulted in the traveling museum exhibit Race: Are We So Different?

At Berkeley, he serves as the associate director of the Haas Institute for a Fair and Inclusive Society, is a core faculty member in the Department of Ethnic Studies and is an affiliated faculty member of sociology and gender and women’s studies. Michael Omi is a recipient of UC Berkeley’s Distinguished Teaching Award, an honor bestowed on only 240 Berkeley faculty members since its inception in 1959. Please join me in welcoming Michael Omi.

Michael Omi: Thank you. Good afternoon folks. It’s kind of this interesting mood lighting in the room. You know, I hope I can keep people awake throughout this. It’s really a pleasure to be here. Nice sight too. You know I come down here to watch musical performances and never thought I’d be on stage. I should have brought my guitar, or something. Next time, I will know that. Well, when Susan Hoffman, OLLI’s director asked if I would speak about the census, this was last July, I wasn’t sure how much interest a talk on the census would elicit. But the census and the politics behind it certainly had been thrust into the forefront of the news with the contentious issue about whether a citizenship question would be added to Census 2020, and the profound effects such a question would have.

So first let me just say something about the census as a whole, and why it’s always been important. Among other things, the constitution does require an actual enumeration of the population in the United States every 10 years, and that means everybody — citizen, noncitizen, etc. It really does provide us with a form, if you will, a kind of a portrait of the nation as a whole kind of portrait of a national accounting, if you will, that provides a kind of collective picture of who we are as a nation.

Very important is the question of political apportionment since political reapportionment is really a zero sum game. By that I mean and many of you know, that the House of Representatives was capped at its current number 435 after the census of 1910, and actually got put into effect about 1911, 1912. What it does is it limits the size of the House. And in so doing transforms, as I said, a reapportionment into a zero sum game. Whenever a state gains a House seat, another state loses a seat. And since 1850, California has generally averaged around three new house seats after each census.

The other aspect of this is federal funding. Federal spending, half of federal spending is allocated on a population basis. That includes Medicare, highway planning and construction education grants, and programs like Head Start. The greater the states under count, the more it potentially can lose. And it was estimated, for example, in the 2010 census that the undercount in California cost California somewhere between 1.5 and $2 billion. So back to this citizenship question.

Last month a federal judge, Jesse Furman, blocked the Commerce Department from adding a citizenship question on the 2020 Census. And just this past Friday, the Supreme Court agreed to step in before the appeals court could rule on the matter and it’s put the case on an unusual fast track. Part of this is because time is of the essence. The census forms are due to be printed in June of this year. So what’s at stake? Well, by one government estimate, millions of people might decide not to participate in Census 2020 and who might those people be? Well, this is a list that might give us some indication of this.

These are people who might be affected by the addition of a citizenship question. What this is from when phrase work is in fact the percentage of residents residing in non-citizen households. And as you could see among race and ethnicity, a great deal of variation here, whereas only 3 percent of whites reside in non-citizenship households. 45 percent of Asians and 46 percent of Latinos do. And this varies according by state as well as you see here in California, 29 percent, Texas, 22 percent. Where there are states with extremely low percentage of non-citizen households. And this includes, for example, West Virginia at only 1 percent.

Well Judge Furman found that if the question were added, Arizona, California, Florida, Illinois, New York and Texas would risk losing seats in the House and that several states could lose federal money. The judge also said that there were questionable motives about commerce secretary Wilbur Ross’ decision to add the question, as many of you know, Ross’ original claim was that it was going to be used as data in support of the Voting Rights Act.

But later it was found out of course that he had very early conversations in his tenure with Steve Bannon. And the point was, how they might, sort of in many respects try to disenfranchise voters in many areas or, at least create some strain on high immigrant states. Furman found that at that much of Ross’ original claim quote was not the real reason for his detention. That’s Furman’s quote. Well, we’ll see what the Supreme Court weighs in on this as well. It could be interesting.

Well, this afternoon I want to focus on the racial and ethnic categories used by the census and hopefully spur some conversation about both their definition and their use. I’m going to go to a case here that appeared in the early 1980s to illustrate, something that I got interested in back then about state definitions, how different states sort of classify by race and ethnicity. And this is the Susie Guillory Phipps case in Louisiana.

What happened in 1977, Susie Guillory Phipps was then 43 years old, found herself in need of a birth certificate in order to process a passport application and she couldn’t find her birth certificate and she never had a passport before. It was the first time she’s going to leave the country. So in fact, she goes down to the Louisiana Bureau of vital records and believing all her life that she is white, imagine her surprise when a clerk at the division of vital records shows that on her reissued birth ticket, she’s going to be designated as black.

Now, what’s interesting about this, well first I’ll give you a quote from Phipps that I always loved. She said, “It shocked me. I was sick for three days. I was brought up white. I’m married white twice.” Now at issue was interestingly a 1970, I want to be clear about that, I’m not talking to 1790 or 1890, 1970 Louisiana state law that allowed for anyone with more than 1/32 black blood to be legally defined in the state as black. And Susie Guillory Phipps had long generational family history in the Lake Charles area in Louisiana. And in fact the states genealogical investigations claimed that she was 3/32 black. There’s a longer part of the story. I won’t bore you with some of that. We could talk about that later in the discussion if you want it. What was interesting about that claim, for example, was the fact that she took that case all the way to the Louisiana Supreme Court and the court ruled that in fact the state could impose the definition of who is black. It was then brought to the U.S. Supreme Court. The U.S. Supreme Court refused to hear it, thereby validating Louisiana’s law. But what happened after that is even a longer saga to this case.

But the point is the designation then of racial categories and the determination of racial identities, really no simple task. And over the last several centuries, it’s provoked numerous debates in this country, debates about, you know, natural and legal rights, over who could become a citizen and indeed who could marry whom. It’s worth remembering that it wasn’t until 1952 that race could no longer be used to deny a person the right to become a naturalized citizen of the United States. And as most of you know, it wasn’t until 1967 in the Loving decision that the Supreme Court invalidated laws which prohibited interracial marriage, and thus the racial and ethnic categories, interestingly in the United States, have historically been put to these purposes and they’d been shaped by the particular political and social agendas of different times.

I’m going to show you something, a chart from sociologist Reynolds Farley. This is a chart of the racial terms that have been used in the 20th century. And what’s interesting about this is that almost every census, we got different racial and ethnic categories. You could see here that at the beginning of the century there are very few racial categories and by the 1980s, 1990s, there are multiple categories of folks being added, a result of different kinds of claims making and politicking around this.

There’s a couple of things which are interesting about this. Of the three consistent racial categories throughout the 20th century was white, Chinese and Japanese. Interesting as labels. I mean black was often black, Negro. It goes off into different tangents, but those are the three specific categories. The other one which is of interest is that Mexicans were regarded as a category, a racial category only once in 1930 and then taken off the census in part because of the opposition of Mexican Americans here in the United States. But also the Mexican government intervened. At question here was whether or not Mexicans would be racialized by the state in particular. And I’ll tell you what happened after that.

Many of you may know this, that the citizens in the territory we took over from Mexico, after the Treaty of Guadalupe Hidalgo in 1848, could elect to become citizens of the United States. And that put them in a very different kind of legal category, and resented in fact having them being seen in the Mexican category. The other ones that are kind of curious here just in passing is that, we have the term Hindu in 1930 and 1940, which was meant to capture the number of South Asians, particularly Asian Indians in the United States. But it’s kind of this way in which a religious category becomes seen as a racial category as well. It’s even more ironic because the largest number of Asian Indians in the United States during that time were in fact Sikh and secondly, Muslim and third, Hindu. But nonetheless, we’ve got this Hindu racial category there.

Now, many of us may be familiar with, something which is called the Office of Management and Budgets, statistical directive 15. The history behind this was that in 1977, the Office of Management and Budget convened a number of federal bureaucracies, you know, the Veterans Bureau too, you know, Agriculture Department, whatever, in hopes of having consistent record keeping with respect to racial and ethnic data. And part of this was precipitated by the passage of significant pieces of civil rights legislation.

So what they wanted to do was that they noticed that different departments, different units were coding people very differently with respect to race and ethnicity. So let’s have some consistent categories and definitions. That was in 1977. It was revised again in 1997 slightly, but many of these definitions we inherited from this period of 77. What you should notice that these definitions for the most part rely on some notion of original peoples inhabiting a specific geographic area, which is always subject by the way to a lot of anthropological debate.

The other thing is that there’s only one explicit racial identifier here and that is for the term black, where a black is referred to as a person having origins in any of the black racial groups of Africa. All right. Yeah. What are the black racial groups of Africa? But there’s no parallel construction. A white person is not defined as a person having origins any of the white racial groups, but a person having origins in the original peoples of Europe, the Middle East or North Africa.

There’s other curiosities with this. The Latino, Latina category, Hispanic category is not seen as a racial one at all. It’s an ethnicity one and in fact it’s the only ethnicity the Feds are really interested in whether one is quote unquote Hispanic or non-Hispanic. And I’ll go back over that to show you the confusion with that. The other one that’s interesting is that not only do you have to be an indigenous person, a person having origins in any of the original peoples of North and South America, but you have to maintain tribal affiliation or community attachment. We don’t require anybody else to have that as a kind of stipulation for them having that significant coding on that.

The other interesting one which I’ll just mention in passing, if you want to ask a question about it later, you can, is that in 1997, the Native Hawaiian and Pacific Islander category was separated from the Asian one. These used to be one category. And since 1977, we’ve only added a new category for Native Hawaiians and other Pacific Islanders. The result of a political debate and controversy by the way. But, native Hawaiians and other Pacific Islanders constitute less than half of 1% of the population in the United States. So it’s kind of interesting how these definitions are being utilized and defined. So originally serving to provide consistent categories for use by federal agencies, the directive has had this unintended consequence of really influencing the very discourse, the very categories of race, which we use today. In many respects, they sort of constitute the main food groups of American multiculturalism, if you will.

Now, here are the questions I’m going to show you and many of you took the census in 2010, I’m assuming these are the 2010’s questions on race and ethnicity. And you’ll see there’s two separate questions and the regular census form only asks about 10 or 11 questions. And two of them are about race and ethnicity. First is this person of Hispanic, Latino or Spanish origin then listing various subgroups but also write-in box.

Then next, what is this person’s race? Also, note this for the census Hispanic origins are not a race, not races, right? Then you’re supposed to come back down here and it’s white, black. You see here a number of Asian categories over this way, a number of Pacific Islander categories, and finally the some other race category. All right? You might want to think about where you would locate yourself in this and how you would fill this out now. I’ll quiz you later.

Now, here’s one issue that has come up with respect to this is that sometimes these categories aren’t very meaningful to the very people they purport to represent, and I’ll give you an instance of this by looking at the Latinx population. In the last four censuses, it’s been estimated that 40% of Hispanics, Latinos in the United States don’t fill out the form the way the federal government would like to see you fill out the form, 40% don’t. What they do oftentimes is skip both these and check off some other race and write in that they’re Guatemalan. They write in that they’re Peruvian, and so forth because they don’t understand how they were supposed to do this.

How the census thought it would work is say if you’re a dark skinned Puerto Rican, you’ll that you would check off that you’re, yes, I’m Puerto Rican and I’m black. Or you’re a light skinned Argentinian that you would check off that you’re some other Hispanic, you’re Argentinian and that you would check off white. Folks didn’t get that. 40% don’t. And it seems to me that 40 percent of the sample, people you’re trying to reach don’t fill it out, right? Something’s wrong with the question, but that continues to plague what the Census Bureau wants to deal with that question. I’ll show you this. This was a survey that was done by the Pew Research Center in 2015. And it’s sample shows that, well over two thirds of Hispanic adults say that being Hispanic is part, both a racial and ethnic part as a reflective of both a racial and ethnic background as opposed to a specific ethnic background. So there’s overwhelming support for seeing these things as combined in ways that, kind of elude how people confront the census and how the census deals with them.

Now what I want to emphasize is that there’s a real politics to, in fact, all of these categories on the census. And one interesting place to talk about that is the politics of the multiracial category because for nearly a century, the census has assumed that each of us, each individual possesses a singular mono-racial identity. Whereas it’s interesting, earlier census enumeration schedules before the 20th sort of recognized some mixed race individuals. The 1890 Census lists mulatto, quadroon and octoroon in addition to the main categories of white, black, Chinese, Japanese, and Indian. Now what’s interesting is that these racial mixed race categories, which are based on quote unquote percentage of black blood, these mixed-race categories eventually disappeared from the census. But what remained in practice was kind of a one drop rule of racial descent that clearly distinguished whites as a pure category from others, including multiracial individuals.

As late as the 1960 census, for example, census takers, were instructed for persons of mixed white and non-white races to report the race of the non-white parent. All right, so the default is you’re white and other you get coded as other right. Now beginning, you know, actually I want to mention a little sidebar on this too. What was interesting, I was told that some multiracial people have certainly ignored census distinctions to fill in only one circle on the race that the person considers himself herself to be by sometimes marking two or more boxes. And since in fact, census scanners were designed to read only one marked box, I was told that oftentimes such persons ended up as monoracial depending on which box they filled in, darker or not, lighter. So, you had to be really consistent on how you were trying to subvert the process.

Beginning in the 1990s, both the Association for Multi-Ethnic Americans, the AMEA and Project Race, great acronym, stands for reclassify all children equally, actively lobbied for a separate multiracial category on the census. Alright. And many of this was being spurred on in school districts where the reporting of schoolchildren and particularly the dramatic demographic growth in the number of quote unquote mixed race, multiracial children, made some of this kind of racial keeping very problematic.

Now, what’s interesting is that I saw this debate unfold in some hearings in Washington D.C. It was interesting that a number, when this issue came up of having a multiracial box on the census, that many established civil rights organizations such as the Urban League, the National Council of La Rasa, Asian Pacific Islanders For an Accurate Census Count oppose the multiracial category. And to some extent these groups feared a diminishment in their numbers and worries that a multiracial category would spur debates regarding the protected status of groups and individuals. According to various estimates, could be that maybe 75 to 90 percent of those who checked the black box could potentially check the multiracial one if it were an option. Not that they would, but the potential existed.

And along with a possible reduction in group numbers, many civil rights organizations argued that the existing federal civil rights laws and programs were based on these memberships in defined racial and ethnic groups. And that from the addition of a multiracial category would make it difficult to assess the salience of mixed race identity in relationship to laws and programs.

What happened is after several years of intense debate, the OMB, the Office of Management and Budget decided to do something nobody wanted to do. The advocates wanted the multiracial category. Major civil rights organizations didn’t want it. What they decided to do was to allow individuals from Census 2000 on to be able to mark more than one box. But it does really interesting things for statistics. Say you’re a black and Korean individual, you would get marked as part added as part of the black population of the United States and as part of the Korean population United States. Therefore leading to something where in fact we’re over. You know, it doesn’t round out to the kind of 100 percent we’d like to see. This is how some of the statistics have to be presented to show people who marked only one or one in combination of others.

Well let me try to bring you up to date about where some of the census forms exist now and then I want to make some political claims about the importance of looking at racial classification. In 2017, actually it was in February, two years ago, the U.S. Census Bureau released the results of what were called the 2015 national content tests. And the census was conducting a lot of focus group interviews. It was using different samples to see how people would respond to maybe different formats with respect to race and ethnicity.

They were also responding to the fact that the nation, as we all understand, has undergone profound demographic change. That there’s been an increased complexity in the immigration flows to the U.S. That in fact, many of these racial and ethnic categories were quite fluid. And in fact there was a lot of widespread campaigns and lobbying for changes in the questions and categories.

I’ll give you an example of that. In 2010, there was a very well organized campaign among Iranians, to tell them, don’t check you’re white, go to some other race and write down Iranian so that there would be a more precise kind of statistical portrait of who Iranians were.

Now they came out with this form that’s called the optimal elements of that test. And I was convinced if you asked me like six months, nine months ago, I thought this was going to be our next census, the racial and ethnic categories that we would use for Census 2020, wrong. But you know, I mean, I’d been wrong a lot on this.

Here’s the interesting features of this. Among other things, it combines the Latino Hispanic question as part of a race question. It’s not two separate questions. It’s collapsed into one. Okay. And they thought that would sort of minimize the ambiguity and the problematic nature about how many of the Latinx population was filling out the census.

The other part of this, which is interesting is that for the first time it allows more details on who is white and who is black. Because before you just check it off, but now you’re able, there’s some other sort of possible subgroup wants to be, you know, French or Polish, among African Americans is to be Jamaican or Haitian or Somali or Ethiopian. It allows you to check off some key examples but also to write in what you want to write in. And neither of those things had been available to whites or blacks, in previous censuses.

Now the other interesting one is the addition that’s been lobbied for 20 years of a Middle Eastern or North African category that would include people who are, you know, Iranian, Egyptian, Moroccan, and also to be able to write something in here too, including Israeli, Iraqi. It’s very interesting how this Middle Eastern category was being presented. The others are pretty much like they were. The main things are, allowing a much more detailed breakdown of who are whites, who are blacks. The combining of a Hispanic Latino question and the addition of a Middle Eastern category. Right. I was convinced that was it.

As I said, I was wrong and it looks like this is what we’re going to get for 2020. What it does is it creates the two questions again. So there’s a question around, Hispanic, Latino identity and then what is your race? You could mark one or more boxes and print origins. It does allow the option of printing origins though still for whites and blacks. This part is pretty much like it looked like for us Census 2010. And what’s gone is the Middle Eastern category once again.

And we could talk about this, there’s been some interesting politicking about the Middle Eastern category over the past 20 years to where some folks wanting it at certain points in time to record disparities or hate crimes. At other points in times, fears among parts of the Middle Eastern community that it could be used to profile them for selective enforcement efforts. So there’s always these questions that exist within that.

The point is is that there is a definite politics to these definitions and categories. They also reveal the kind of social constructiveness and the fluidity of race. And in many respects these racial and ethnic categories can really be seen as a kind of effects of political struggle over their meaning and definition and in turn these categories can have political effects in sort of setting the stage for the next round of debates about classification.

This gets me to something I want to talk about, which is about a racial classification and colorblindness. And, and in many respects this is trying to get at the political meaning behind why we racially classify, the reasons we do it, and its significance and its impact in many respects. Now, I won’t go into this too much. You know, many familiar with this, the ideology of colorblindness, that many people believe that the goals of the civil rights movement had been achieved and that racial discrimination is a thing of the past and that we are now truly a colorblind society. Colorblindness really denies that race or concepts or race inform our perceptions, shapes our attitudes and influences our individual collective and indeed institutional practices.

And oftentimes it’s being said that colorblindness is the belief that the most effective antiracist gesture policy or practice is simply to ignore, ignore race. Now obviously this says as a different tinge in the Trump era as, you know, where before we used to think about coded racial politics, but what a colleague of mine on campus, Ian Haney Lopez defines as dog whistle politics. And it’s really true that Trump has replaced a dog whistle for a bull horn. And in many respects, you know, the explicit racist appeals have much more currency now in kind of normal political discourse.

Well for over the past, I would say 20, 25 years, there’s been this concerted attempt by mainly political conservatives to ban the collection of racial demographic information. And that government policies that utilize racial categories. This inclusion exclusion has been criticized for promoting color consciousness and subverting the ideology and practice of colorblindness.

Many of you may recall this, that in 2003 California, had a Proposition 54, the Racial Privacy Initiative, which was sponsored by former University of California Regent Ward Connerly, who is the architect of the Anti Affirmative Action 209 in the state along with Michigan’s Proposal 2. And, what the Proposition 54 sought to do was to add to the California Constitution to make it, illegal or the state shall not classify any individual by race, ethnicity, color, or national origin in the operation of public education, public contracting or public employment. I like this quote. Ward Connolly believes that, getting rid of the data is important to move us towards a colorblind society. And he wrote this argument, which was in the California state voter pamphlet. He says, “Dare we forget the lessons of history. Classification systems were invented to keep certain groups in their place and to deny them full rights.”

Well, I think a problem here is that Connerly fails to make a historical distinction on the use, the use of racial classification in the pre- and post-civil rights era. Prior to the passage of civil rights legislation in the 1960s it’s true census categories were utilized to really politically disenfranchise and discriminate against different racialized groups from prohibitions on naturalization rights to setting quotas for the 1924 National Origins Immigration Act. Many of these kinds of categories were used to circumscribe the political, social and economic rights of specific racial minorities.

But after passage of civil rights laws, many of these categories, much of the census data was used to discern the patterns of discrimination that were being practiced by individuals, but also by businesses, by schools, by political institutions, against people of color and other marginalized groups. And in fact, data by race became absolutely critical for the enforcement of every civil rights law passed since the 1964 Civil Rights Act. And thus the civil rights revolution, if you will, or era marked an important shift in the use of these kinds of racial categories from a tool to exclude specific groups, exclusion, to one to ensure inclusion and a much more robust representation of groups.

Now that said, it still comes up much about whether or not we should simply abolish racial classification. And I would argue that we can’t simply abolish forms of racial data collection in the United States because the reality is is that without some form, some form of classification, some form of racial data, we’re really unable to observe institutional patterns of inequality in the United States.

The Institute of Medicine had this enormous study on equal treatment that found that even if you controlled for insurance coverage an ability to pay, there were stark inequalities by race in terms of preventative care, diagnostics and treatment no matter what the disease in question was. In one of the the studies they cite, a study of 1.7 patients, blacks received major therapeutic procedures less often than whites in 37 out of 77 medical conditions. And although another example, although, we passed a Fair Housing Act in 1968, which made it illegal for lenders to use race as the basis for lending decisions, it was not until we had data collection laws in the 1980s that lending patterns, home loans, home mortgage loans could be clearly discerned. And of course, what was found is that the loan rejection rates were twice as high for blacks as they were for whites who had similar income portfolios.

Now these examples illustrate the need to maintain some form of racial classification and racial data. But I want to say this as well — that this does not mean we could ever have accurate, precise or scientific categories. You know, it’s an elusive quest to find a category that’s going to be, be able to be enshrined forever. Racial categories and the definitions and meanings we impart to them have changed over time, but they cannot be simply dispensed with because in a racially stratified society, social concepts of race, social concepts of race matter. And in fact we need to examine patterns and trends to be able to really inform policy to inform how we confront these disparities.

Now that said, in the Trump era, there’s a lot of dark clouds on the horizon with respect to data collection. Again, I’m going to show you an early example of this, which I don’t think, finally got passed, but there’s been different attempts to do this. This is a bill in 2017 that was sponsored by Senator Mike Lee and Marco Rubio. And it’s called the Local Zoning Decisions Protection Act of 2017. And one of the things that it’s explicitly prohibited, which is very interesting, is that, it said that no federal funds could be used to design, build, maintain, utilize or provide access to a federal database of geospatial information on community racial disparities or disparities in access to affordable housing.

Let me deconstruct that a little bit. In other words, what was very important to think about a housing policy was in fact the maintenance of some sort of federal database that would be able to tell you, where there are racial disparities in residential communities, or disparate access to, in fact, people trying to secure affordable housing.

You know, what the parallel with with this is on climate change. You know, you get rid of the data on climate change and then maybe the problem will go away. Yeah. And in the same token, it’s saying, get rid of the data about race and housing and maybe the racial disparities will go away. What will go away is our ability to actually see the depth, the impact, the significance of those changes without any form of data.

I’m going to end on this quote. This was from Ken Pruitt who was the director of Census 2000. He sent me an article on the Proposition 54 debate, that Racial Privacy Initiative, and Ken Pruitt says, I love this quote. He says, “I’m happy to join Ward Connerly and welcoming a colorblind society, but I don’t want to be blindfolded as it arrives.” And, he’s right. So thank you.

Oh, I should just take some questions and it’s hard actually to see in this arena, I got to say, I’m going to go to the second row here. Yes.

Audience 1: Can you comment about the citizenship question, the debate and where you think it’s headed?

Michael Omi: Yeah. So, were you here where I began with the citizenship question or I did mention the citizenship question at the beginning of my talk, particularly of Justice Furman’s, regards to the thinking that Wilbur Ross is being disingenuous about why in fact the citizenship question was being added. Ostensibly, he said to support and make data available to secure the Voting Rights Act. But it really was clear from discussions with Steve Bannon and others that the ulterior motive was to try to disenfranchise groups of people and that would change our political landscape. In fact, to be able to do that, as well as penalize some high-immigrant states that in fact, you know, the Trump administration had some grievances with.

Where that is. So the last Friday, as I said, the Supreme Court fast tracked the case, which is extremely unusual. So in other words, the Federal District Judge Furman’s decision will not go now to an appeals court. It will be heard directly by the Supreme Court. And in fact, that’s because, as I mentioned, the census forms are, are due to be printed in June. So that decision has to come down fairly quickly.

Where is it? You know I’m going to hedge, my bet here, I was wrong about all these census forms. I’m sort of gun shy to talk about something else. But, I worry about that. I actually think that given the current stuff of the Supreme Court and some earlier remarks that Neil Gorsuch and some others have made about, the ability in this case of Wilbur Ross to initiate these changes that it could indeed go through. It’s going to be an interesting thing. I think what’s key here is it could be fairly split. And, here again, I’m not sure how people are going to lie on that, but many people,

I know why I say that is because Neil Gorsuch didn’t find that Wilbur Ross was being disingenuous or you know, that it wasn’t proved that he had a darker, quote on quote purpose in mind and doing that. And I thought that was a flag to me that, you know, if you don’t believe he did it for reasons to sort of disenfranchise folks, that mandate or that instruction to the census may go through.

Audience 2: Yeah, so you alluded to the concept of hypo-descent the legal definition of blackness being 1/32 black blood. So I’m wondering if maybe you might be able to comment on how the concept of hyper-descent in Latin America maybe informs this notion of being Latino or Hispanic as an ethnicity versus a race.

Michael Omi: Oh, interesting. So should have to say what is a hypo-descent is referring to descent rules that when you have in a status hierarchy, a higher status group and a lower status group, any intermixture assigns you to the lower status group. But you were talking about hyper-descent.

Audience 2: Yeah. So I seem to remember reading somewhere just this, that hyper-descent in Latin America, this notion that any small amount of whiteness confers status on individuals in Latin American society, that this has somehow a relation to the whole notion of being Hispanic as an ethnicity on the census.

Michael Omi: Got It. No, I think there are two separate things. And let me comment on both. On the one hand, it’s very interesting to do a comparative analysis on how racial categories are defined in other parts of the world. And certainly in Latin America it’s very interesting given the kind of real hybridity, you know, the mestizo mestiza consciousness among Latin Americans there.

But also you have something else which is a kind of, there is a pigmentocracy of lightness. What’s interesting to me is that you could be siblings from the same set of biological parents in Brazil and be assigned to two separate racial categories because of colorism in many respects. That kind of illustrates a different understanding of race than we have in the United States. But you know, the notion about all white blood, you know, sort of the lightness. It’s also class stuff. You know, there’s an expression, I believe in Brazil, but maybe other Latin American communities, is that money whitens such that class and class privileges has the ability to transform your racial location.

Now, I think that’s different than the Hispanic thing. Here it was sort of defined, not being seen as a racialized category. It was lobbied for, you know, the creation of a kind of Hispanic category. And, it was done before a real diversity among a kind of Latino, Latina population too. I mean, there’s real shifts going on with respect to this and what some people didn’t want about collapsing the Latino category as part of a race category was that they wanted to be able to discern — this is colorism thing again — whether or not people who were dark-skinned or you know, Latinx folks were treated differently than white folks in terms of, you know, everything and that might be more negligible unless you were able to record them by race as well. So I think it was two separate things. Yeah.

Audience 3: You know, on a personal basis for a lot of people that have had their or whatever, they see that they’re mixed in all kinds of ways and the census can almost force an individual to deny something unless they check it all and also choose what they want to identify with, given certain percentages. The other piece of that just personally is Mexican Americans are mostly mestizo anyway, they’re indigenous and Spanish. So that whole mix creates a lot of confusion as well.

Michael Omi: Absolutely. It does. Let me deal with the first question. I talk another thing about the genomic sciences and race, too, and in which you’re looking at ethnic ancestry as well as criminal forensics and drugs around. I won’t go into it now, but it’s been very interesting since, during the Clinton years, the early 2000s, when the human genome was first mapped, there was this big claim about how that shows that we’re all one race, that there are no racial differences. But what’s been really interesting is that the short number of years after that, everybody started thinking about race and genetics as having a particular thing with susceptibility to certain diseases or being able to do, you know, treatment as well as ethnic ancestry. Listen, I won’t go into it, I have colleagues who are better equipped at this, but there’s a lot of problems with the ethnic ancestry testing as currently construed by 23 and Me or It’s based on very limited samples of people. I mean, that could expand and get better, but before, you know, they’d give you these big swashes of who you could potentially be.

And the other thing, you know, to say you’re, wow, I’m 10 percent Costa Rican question is, who’s 100 percent Costa Rican. So there’s real problems in both the sampling and the logic behind some of that ethnic ancestry. And that’s why the census is very clear. They say, we’re not talking about biological, we’re not talking about any of those definitions. We’re talking about social concepts of race.

Now that said, you know, you can put anything you want to put and that’s because we’re under self report now and we not always were. Many of you may be old enough to remember where a census taker came to your door and looked at you and checked off boxes and was only supposed to ask you if it was, if they thought ambiguously, they weren’t sure who you were. But now it’s driven by self report so you can say anything, you could describe yourself in any terms you want to. That’s the way it goes.

Audience 4: You know, this is all kind of confusing to me. I’m a Rodriguez, Rodriguez family lines, been in California since the 1700s. I’ve seen the census reports in my family. Spanish, Mulatto, Spanish and then for about 200 years, white. And then, when I was 40, I found out I’m Hispanic. The Equal Opportunity Department of Labor came to my workplace and wanted to interview employees at random. And so my boss said, “You need to go to this interview. And I said, oh why? Because I’m over 50?” He said, “I don’t think so.” I said, “Because I’m a woman?” He said, “I don’t think so.” He said, “I don’t really know. You’ll have to ask them.”

So when I went, they said, “You’re Hispanic, that’s why you’re here.” And I said, “Well, I don’t know. I don’t think I’ve ever said I was Hispanic. I didn’t ever fill out a form.” My maiden name was Rodriguez. I don’t even know if the company knows that. Is that okay? He said, “They would be in violation if they didn’t list you as Hispanic.” So now I’m Hispanic and it’s okay. I only have to think about this every 10 years when the census comes out. But looking at those things, you know, I still don’t really know how to fill them out. The second example. I thought that might be easier. But you didn’t like that one?

Michael Omi: Well, you mean the one that, uh, which one are you talking about here? Let’s go get back up. The latest one?

Audience 4: Yeah, that would, that one I think I could fill out easier. But I don’t know. The only thing, I thank you, I think your lecture is great. The only thing I don’t agree with you is I really don’t think any of this matters.

Michael Omi: Well, okay. We disagree on that. Yes.

Audience 5: I just wanted to state that I think it’s patently absurd for them to put the those print, for example, German, Irish, blah, blah. My daughter has five different European ethnicities in her. We’re Americans. We’re all a mixture. We’re mutts. We’re not you know, and so what is she supposed to put? You can’t print in all that different stuff. And same with me. I have three. Yeah. And they’re not, you know, they don’t, it says print.

Michael Omi: And people make choices.

Audience 5: Yeah. I understand, but I think it’s stupid. It’s not taking into consideration what we are.

Michael Omi: Well, let me tell you this, that this is another study that I think Jeffrey Purcell or someone did, which is that whatever these examples are have an enormous effect on what people put down. Because this was derived from I think the American Community Survey where it gave you examples and then every time there were different example, oh no it’s a long form of the census, which we don’t have anymore. It would have examples of your specific ethnicity. And you know, one time they dropped off French and the French population in United States fell by 20 percent. And it didn’t. But people didn’t see French then thought, “Oh my god, that’s right. My grandmother was French and write stuff on.”

So yeah, even the examples themselves, have an undue influence on how people identify. But you’re going to get that and you’re going to get like siblings do different things, you know, they’re going to put different things down and people do put different things down on the census. And then some people who are trying to capture this. I had a graduate student, she was Bangladeshi, extremely dark skin, grew up in Texas. She showed me her birth certificate, says she’s white. And she said that, when she was born, the nurse or whatever who was filling out the form with her parents, didn’t know how to capture who they were. So she just thought white. Yeah.

Audience 6: Wondering what, if any effect does this data, racial data, have impact on the reapportionment?

Michael Omi: This data does not explicitly. But it does deal with the reapportionment debate with respect to the addition of a citizenship question, whether or not a citizenship question would have a kind of depressing effect on the number of noncitizen households responding to the question. And that would reflect on the reapportionment. That’s why Judge Furman thought there was like five or six states that would be affected in a reapportionment debate should that question be added. But it’s not directly based on a kind of race stuff.

That said, there’s all this stuff going on. Many of you are familiar with at the state level around voter suppression and particularly the ways in which both past histories of Gerrymandering, but also a contemporary stuff has as led to a significant taking people off the rolls, many of whom are people of color. What’s been very interesting for example, is places like Florida passed a law about whether or not ex convicts, ex prisoners can in fact vote or not. Those are extremely important kinds of cases about how to expand voter roles. But what we’re faced with for the most part is a kind of a legacy of voter suppression policies. And those more or less have racial meaning. And if we didn’t have racial data, we wouldn’t be able to figure that out, about who is being suppressed.

Audience 7: How can you do any kind of studies year to year with this constant change of these categories? How can you have any accuracy and comparison?

Michael Omi: That’s an extremely important question because that’s what, yeah, there was a person here in demography that hit me up on that. She said, “ell, how can I do longterm studies if what I mean by this category changes over historical time?” But that’s a fact that you have to grapple with, that the categories themselves may change. I mean, some have a role to permanence to them, to historically about who is African American or black in the United States is one. There’s mainly the kind of addition of new groups, whereas somebody who’s Iranian may have been coded as white for a number of years, all of a sudden becomes Middle Eastern, so there’s kinds of more complexity in addition to that.

But it’s true. It does disrupt to some extent the kind of time-series data. But I actually think for the most part it doesn’t. I mean, in early censuses as well, in 19th century, think about this. We understood people like Germans to be a race. We understood the French to be a race. It wasn’t consolidated in a kind of homogeneous category of whiteness. Think about the Irish, think about the kind of discrimination and just outright hostility directed to the Irish at an earlier point, which then leads to thinking about, well, what happened? And we have historians like Noel Ignatiev who wrote a book called “How the Irish Became White” as a way to talk about that kind of historical transformation and its broader meaning.

Audience 8: Are there penalties for lying on the census form for claiming to be a citizen, for example, when one isn’t?

Michael Omi: Not that I know. No. Although the fear is, is that it would have an overall chilling effect on people filling out the census in the first place. So, but you’re not fined for directly lying on the census, I don’t believe. I had a friend of mine who was a community-based organizer in Chicago who is in fact white and every census he puts down he’s black to increase the number of blacks in the Chicago metropolitan area. So, you know, maybe they could get him at some point. But, no, I think the point is, is that people won’t fill it out.

Now the Census Bureau, the Office of Management and Budget, the Commerce Department is very clear about that we would never use this data for like ICE raids or to figure out, you know, but who believes that, you know, is one thing. And I’ve got to say at the same person, Ken Pruitt, when he became director for Census 2000 had to issue an apology on for the Census Bureau for something it did 50 years ago, which was the use of census data to figure out where Japanese Americans were in order to evacuate and incarcerate them during World War II. And he said, that was a blot. We shouldn’t have done that. That data should never been released, but it was, and I’m thinking, “So what’s to prevent anybody else from using this data in a kind of a malicious way?”

Thank you so much. Thank you.