Promising an enlightening journey through the historical and contemporary significance of the U.S. Census, I'm joined by esteemed scholar Dr. Dan Bouk - professor of History at Colgate University and author of the book Democracy's Data: The Hidden Stories in the U.S. Census and How to Read Them. This episode unearths the nuanced power dynamics and biases inherent in the census process. With a focus on the 1940 Census, a task monumental in scale, we uncover the detailed process of transforming raw data into a compelling narrative about who we value as a society.
We investigate the poetic yet complex process of conducting the census, shedding light on the intricacies of data literacy and the potential pitfalls of interpretation. Dr. Bouk offers valuable advice for researchers and analysts, emphasizing the importance of recognizing and conveying uncertainty in statistical data. Moreover, we explore how this counting of people creates narratives about societal importance, resonating particularly when considering marginalized groups and queer communities.
Unmasking the suppression, survival, and blossoming of marginalized communities through the lens of census data, we endeavour to understand how this vital tool impacts the power distribution among states. We analyze the effects on American democracy, considering the challenges and potential of an accurate and inclusive census count. In conclusion, we reflect on the immense potential of the U.S. Census, not just as a data gathering tool, but as a means to shape a more representative and united America.
Recommended:
Democracy's Data: The Hidden Stories in the U.S. Census and How to Read Them - Dan Bouk
Mentioned:
The 272: The Families who were Enslaved and Sold to Build the American Catholic Church - Rachel L. Swarns
Close to the Machine - Ellen Ullman
Thinking Like an Economist: How Efficiency Replaced Equality in US Public Policy - Elizabeth Popp Berman
The Wandering Mind: What Medieval Monks Tell Us about Distraction - Jamie Kreiner
-------------------------
Follow Deep Dive:
Instagram
YouTube
Email: deepdivewithshawn@gmail.com
Promising an enlightening journey through the historical and contemporary significance of the U.S. Census, I'm joined by esteemed scholar Dr. Dan Bouk - professor of History at Colgate University and author of the book Democracy's Data: The Hidden Stories in the U.S. Census and How to Read Them. This episode unearths the nuanced power dynamics and biases inherent in the census process. With a focus on the 1940 Census, a task monumental in scale, we uncover the detailed process of transforming raw data into a compelling narrative about who we value as a society.
We investigate the poetic yet complex process of conducting the census, shedding light on the intricacies of data literacy and the potential pitfalls of interpretation. Dr. Bouk offers valuable advice for researchers and analysts, emphasizing the importance of recognizing and conveying uncertainty in statistical data. Moreover, we explore how this counting of people creates narratives about societal importance, resonating particularly when considering marginalized groups and queer communities.
Unmasking the suppression, survival, and blossoming of marginalized communities through the lens of census data, we endeavour to understand how this vital tool impacts the power distribution among states. We analyze the effects on American democracy, considering the challenges and potential of an accurate and inclusive census count. In conclusion, we reflect on the immense potential of the U.S. Census, not just as a data gathering tool, but as a means to shape a more representative and united America.
Recommended:
Democracy's Data: The Hidden Stories in the U.S. Census and How to Read Them - Dan Bouk
Mentioned:
The 272: The Families who were Enslaved and Sold to Build the American Catholic Church - Rachel L. Swarns
Close to the Machine - Ellen Ullman
Thinking Like an Economist: How Efficiency Replaced Equality in US Public Policy - Elizabeth Popp Berman
The Wandering Mind: What Medieval Monks Tell Us about Distraction - Jamie Kreiner
-------------------------
Follow Deep Dive:
Instagram
YouTube
Email: deepdivewithshawn@gmail.com
The tragedy. There is not a tragedy for us now in which we can see back and start to recreate these strange and complex relationships. The tragedy is to think that this was yet another official statistical manifestation of the erasure of queer life that was happening constantly at this time, and that's a tragedy to understand. It is also, though, an important thing for us, a way that we gain knowledge about the past. By thinking about all of the parts of the data, it helps us really understand exactly how it was that queerness both survived, persisted, was honored by the people who were living it, and then was suppressed and pretended to be non-existent by officials with the power to make the final statistics.
Shawn:Welcome to Deep Dive with me, s C Fettig. Today I'm focusing on data, specifically census data how we count people, the profound implications of this counting process and how it shapes the very fabric of our democracy. The United States Census is a vital tool that has evolved over centuries to offer a snapshot of our nation. The Census, seemingly just a statistical endeavor, has far-reaching consequences. It reveals the story of how we, as a nation, value some groups and individuals over others, and it holds the key to understanding representation, funding and the very essence of our democracy. The impact of the Census extends beyond mere numbers, though. It informs policy at every level, guiding the allocation of resources and influencing the decisions of policymakers. Yet beneath the surface lies an intricate web of challenges and biases that we must unravel to ensure an equitable and just society. The goal today is to not only uncover these complexities, but also shed light on how we can improve our methods of collecting and utilizing census data. We Americans aspire to harness the potential of this invaluable tool to bolster democratic ideals and create a more inclusive and representative society. It's, in fact, the very reason we have a Census.
Shawn:My guest today is an esteemed scholar in the field, d Dan Bouk.
Shawn:Dr. Bouk is an associate professor in, and chair of, the Department of History at Colgate University. With a deep-rooted passion for history and expertise in the quantitative analysis of historical data, d Bauc has made significant contributions to the understanding of how we count and interpret populations and bridges the gap between history, data, science and policy, offering invaluable insights into the steps necessary to attaining a more equitable, just and representative future. His most recent book, democracy's Data the Hidden Stories in the US Census and how to Read them, anchors much of our discussion today. The conversation we have illuminates the intricacies of census taking and sheds light on its historical significance and the way it has and continues to influence the lives of individuals and communities in America, as well as our very democracy. If you like this episode, or any episode, please give it a like on your favorite podcast platform and or subscribe to the podcast on YouTube. And, as always, if you have any thoughts, questions or comments, please feel free to email me at deepdivewithshawn@gmail. com. Let's do a deep dive, d B. Thanks for being here.
Dan:How are you? I'm great, Shawn. Thank you for having me.
Shawn:Absolutely so. The catalyst for this conversation, and why I'm excited to have you here, is your book Democracy's Data the Hidden Stories in the US Census and how to Read them, which is a deep dive into the 1940 census and what we can glean not just from the data, but how that data was collected and how it reflects America and our policies. And before we jump in, I do want to say that this book is really remarkable, and this isn't just a platitude. I think this book is a great drama in a way, a really readable peek into history and US attitudes and policy, and also a digestible education on data and why it matters. So I'm excited to have you here.
Dan:I love that you called it a drama. That's very satisfying. Thank you for having me.
Shawn:Okay, so first things first, though, to help the listener understand why you focused on 1940 census data and not any more recent years, why 1940?
Dan:So you could perfectly reasonably say that every single census is in fact its own very significant drama. I mean, if that sounds like a strange thing, actually, you just need to kind of repurpose or think about this again. The project is find every single American all in 1940, it's 130, some million, now it's 330, some million Find each one of them and basically try to touch them almost and make sure that they actually exist, all within a space of like a couple months. And suddenly you really think like whoa, actually that's like scaling Mount Everest or like trying to find the North Pole. It's like a curriculum thing. So it could have been any. I think you could have reasonably done any census.
Dan:I started with 1940 because it was the when I started writing the book.
Dan:It was the most recent census for which we had all of the manuscript material, which is to say that when a census is completed, the Census Bureau immediately, as fast as it can, starts producing statistical tables full, you know, all the reams of numbers explaining how many people are there in each state, and then finer geographical levels and then broken down by a host of different demographic categories.
Dan:That all happens. It comes out right away, but what each person is supposed to have said to an enumerator, or what they typed into an internet self-response form. Now these days, all of that is held private and completely confidential from also anyone else in the US government for 72 years, and so when I started this book, the 1940 census was the only one that was really available, and then that meant that it was both a very modern census but one that I could like read deeply, and I liked the idea that I was starting in the same place that many people who do their own family or community histories also start, because if you're beginning a genealogy, you would also often begin with that most recent census and then kind of build back from there.
Shawn:I mean if this was new information for me. A lot of governments state, local and federal will protect certain pieces of information or certain types of information for periods of time, so it can't be released until X number of years have passed or a certain date. We think about the Kennedy assassination or certain things related to the Nixon investigation etc. And those kind of make sense right, like to some degree they're protecting either investigative methods or they're protecting investigators or subjects etc. It's not entirely clear to me why the details of the census are under lock and key. I think you said you say in the book for 72 years, so each census correct, that's right. This portion of it is under lock and key for 72 years and it's not entirely clear to me why.
Dan:So I think there's a couple of different answers that are both satisfying Partly your right to have that instinct that it's not immediately obvious or necessary. So throughout much of the early history of the US Census that there was very little confidentiality in the making of the count, in fact, us Marshals so back then it would be like essentially the law enforcement would be in charge and they would commission other folks to make these counts and then they would be posted prominently in public areas that people could check to make sure that the count had happened properly. Part of the reason it comes to be put under lock and key, as you put it, is that there's a sense that the government starts to ask more and more questions about people and to want more and more information, some of which increasingly might be seen as something that one doesn't want to get out. Or there's a concern, increasing concern that you maybe don't want to tell the government this thing out of concern that then it might be used against you in some way. One really easy way to think about this is in the 1940 census there was a question about income. It's trying to be very controversial, but there also was increasingly a move towards taxing people according to their income, and so the people who make these statistics were reasonably concerned that people might not want to give a certain kind of information or honestly disclose their income if the tax man might then get that information and use it against them. That doesn't even have to be. This will kind of move into my second answer. You don't have to think about this as like people are liars and they want to hide their money from the tax man, although undoubtedly that is true in many cases.
Dan:So in the 1950s there's a series of Supreme Court cases and then a change in legislation that has to do with whether or not a company that submits its information to the Census Bureau about its own profits when it keeps its records, whether those records that it keeps it then has to give up to the Federal Trade Commission to determine whether or not it's part of a monopoly or to do other kinds of regulatory things.
Dan:And one of the things here is the idea is that the Census Bureau wants companies to give it fast data about its operations, because it can then use all these different companies fast data in a standardized form to come up with a reasonably accurate picture of what the economy looks like in a region or in an area, or how what an industry looks like in a particular place, even if those numbers are not precise in the way they would be for an IRS statement.
Dan:And so because by the time you pay your taxes it's four months later than the time that the stuff is actually happening. And so sometimes part of the reason you want confidentiality is because data that's good enough to give you a good statistical picture when it's averaged out over an entire population, is not actually a good and accurate representation of the individual, the individual person, the individual company, and so confidentiality or privacy is useful, because the data double that stands in for you with this bureaucracy is fine for statistical purposes, but if it got out and people used it to try to then do things to you in the present, that would be inappropriate. So it's an argument for privacy.
Shawn:So reading this book tapped into two perhaps conflictual reactions, or maybe feelings for me. And it was gratifying in reading such a careful parsing of some of the unique characteristics of who we people living in America are, but it was also frustrating in recognizing how limited that knowledge is based on what you can get from the census, so limited by a lack of standardization or limited by the pieces of information that fall between the cracks of the information collected, or limited by politics, et cetera. So I'm wondering how was this experience for you, researching and stitching together the stories and seeing the data as it was?
Dan:All right, so I have an answer for this, but I guess I'm a little bit curious about your own explanation of this frustration. Did you wish that? Did it feel like a series of failures, like why don't they, why do they keep not keeping the right kind of information about these people? Or did it just feel too close to home with your own day to day work with all this data?
Shawn:I think it's hindsight primarily that you know when we sit down with data and I'm sure you feel the same that you want like a perfect data set, right?
Shawn:And you want especially if it's time series or if it's data that's been collected across different communities that it's somehow standardized in a way that is useful and can be compared against each other in different contexts and different times, but is also thorough and sensitive. And I think what came through is that the census is a battle, or at least how the census is designed and the process is implemented and how people will be counted and what are the types of things that we will count and how we'll ask those types of questions. That's a battle every 10 years that plays out before it ever lands on our doorsteps, right? And what that means is that there is no standardization. The questions are different, they're speaking to the time and in every instance, you know, from decade to decade there's deterioration. If the question changes, you know it might be asking essentially the same thing, but because we've changed a little bit how we talk about gender, right, we can't compare the answer to gender today to how people might have answered it 100 years ago, and that's just frustrating, right?
Dan:Yeah, Now another. You put it like that. I can. I totally see what you mean. So I think my primary emotion throughout actually a lot of the book, maybe a surprising amount of the book is delight, Like just that the stuff exists at all, there's. I see this whenever I bring students into like an archive of any capacity, like just even in our little university here at Colgate, and you touch documents from 100 years ago. It really feels like time travel and it is, in a way, right Like it is literally the past, sitting there with you and every single time I found a new person, which was a new person on every single one of these sheets, and then every time I thought about what I was really looking at.
Dan:So I guess one thing that the senators should be thinking about is that now many people here who are listening to this, if they filled out a census in 2020, probably submitted their information online and they probably didn't answer a question asked by any human person. They just like typed in a series of responses. In doing so, they undoubtedly shook their heads at least once when they're like we're asked a question where they're like I don't know how I fit in here, or like that's a weird way. So like that part happened. But back in 1940, every single one of these interactions would have been between two people, or like almost every single one where there was a numerator who knocked on a door and who was asking a question, getting a response and then writing down the answer. So when I'm looking at, when any of us are looking at these sheets, we're seeing a mediated response in which this enumerator has heard something, is making judgments, and then is writing down these things. So it's like it is such a fascinating document that here I'm seeing from almost 100 years ago, a conversation that has been perfectly recreated, but then like run through this weird machine that breaks it up into all of these cells, like on an Excel sheet, and answers only in little pieces that are each one a little bit suggestive and yet leave just enough ambiguity that you start to imagine and like wonder what else is happening.
Dan:So I I cannot look at census sheets for more than five minutes before I start speculating wildly, and so a great deal of the book is trying to like hold back that in that instinct. And yet it's also like that's. That's just part of the. What's so delightful about it is the sense that, like these are real people. They are super complicated, so complicated that they don't really fit inside these simple cells that we've constructed for them. And I get this now, this joy of getting to like see how they were stuck in there for this moment, recognizing it might have been kind of pretty miserable for them at the time, but at least something about them was preserved. And now we get to like sit with them and try to think like what was the bigger person, the like fuller person that is represented by these few marks?
Shawn:You know, I think this is kind of fascinating because to some degree, I think this might be highlighting just a difference in how we look at data and by we I mean you and I and how we interpret the right turns in the data, right? So what I was mentioning to you as being frustrating was essentially what flew off the page to me and not necessarily in the book, but in what you were talking about related to the census was just all of the weaknesses, or all of the breaks in the data story along the way that for me, I guess, in the aggregate, look like just cumulative chaos, right? And what you're saying is that these to you appear to be and you can correct me if I'm mischaracterizing opportunities to imagine what was in those breaks or what could have existed there, or how the breaks, the ways that they were bridged, are somehow telling maybe a coherent story. And then what's in that gray area?
Dan:Yeah, I mean that's a wonderful way of putting it like that.
Dan:In the breaks there's is the chance for story, in the breaks is the possibility for us to speculate.
Dan:But in the breaks also is just a way of acknowledging the way life really feels to us most of the time, which is that we don't fit well in standardized categories, even when they're really finally parsed.
Dan:And so I suppose a lot of this is getting down to my own ambivalence about standardization, where I don't know any other way to run a mass society or how to make a more fair society or how to make a place we can all live in together than to have statistics. So I'm actually so I'm for statistics and at the same time I really deeply recognize how much gets lost and how difficult it is to actually turn a person into them. This is, in a way, why I ended up being such kind of a radical adherent to principles of privacy, because I think the only way to square those two that circle for myself is to say we need to allow people to be manhandled in a way to be to turn them into useful statistics, but then we should keep them, those people, as far away from those represent representations of themselves as possible, because they're not accurate representations.
Shawn:I think, now that we've kind of called this out, the different lens through which we kind of interpret this, now I can see how the framing of my next question is absolutely through my lens.
Shawn:So bear with me. One of the characteristics and perhaps weaknesses, I think, of the census data so there it is, perhaps weaknesses you maybe don't see it that way Of the census data that comes through in the book to me is how stories and experiences and conceptualizations, which in end represent the actual lives and histories of real people, is embedded in the processes of census taking. So the data collected, how it's collected, who's doing the collecting, how it's analyzed, all of these things that you've kind of mentioned already, and then also the absence of each of these, so what's not collected, et cetera. So what are some of the processes that generate the census numbers, or at least through the time periods that you were interested in, that have and maybe still do inform the stories the census tells us, and what other processes, if you've given us any thought, if implemented, could tell a slightly different story based on the same numbers?
Dan:Yeah. So one way to answer that is to say that I think actually, any data set, particularly like any large data set, we can understand it as having four broad phases, and when we normally think about, like oh, how is data made, we think of two of them. So we're, like, are pretty comfortable talking about like well, probably at some point there was somebody who designed a set of questions, or decided what the categories would be, or marked out some rows and columns, and then, having decided, like all right, we're going to count the number of apples and oranges, we figure, like all right. So then what the data looks like is when we have a number of apples and a number of oranges in a spreadsheet or printed on a table. In between, though, is like and this is where, like the stuff that I'm, the other stuff that I'm really interested in, comes.
Dan:I think of those moments. The moments we think of most are the places where, like, there's the most order and the most centralized control. So, for the census in 1940, that moment of centralized control meant that a whole bunch of people were drawn together to the Commerce Department, which is the part of the US government that runs the census, and a lot of them were working for government agencies, a lot of them were working for the Census Bureau, a few other folks representing labor, representing philanthropy or universities. They all would show up some big people from big business and they would hash out together these are going to be our questions. And then and today I'm sure you've encountered this limit you can only ask so many questions, like the fundamental limitation for the folks in 1940 was there was literally only so much room on the page, but you also just have only so much patience and so much of the people you're going to be asking questions of. So you do have to make choices, and so you hash this out, you come up with questions. That's the first process.
Dan:Then, having put those sheets together that are blank, you send them off with all these enumerators to go into all these different communities. And this is one of these moments where that chaos breaks loose. And that chaos is. It's dramatic in the sense that it does create an enormous problem for the people who actually want to do the counting, because they're going to get all kinds of weird responses and enumerators even well-trained enumerators dealing with complicated situations, are going to end up putting down on these pieces of paper stuff that doesn't fit into any of the established categories. So, having gone out, you collect data about all the world.
Dan:It's handwritten on pieces of paper, sometimes crawling over the edges and often with things that just don't quite fit, and all of that then gets sent back to Washington DC, to a warehouse in this case, that was rented for the purpose because there wasn't enough space, and then eventually they were able to build a new building.
Dan:They take all this material and this is the third phase, and this is the phase where it ultimately leads to the printing of statistics, and we can talk about this as data cleaning or we talk about it as editing, going through and making sure all of the responses actually fit. One of the acceptable answers putting them on punch cards, pieces of paper where you could punch in the responses, using those to then make tabulations, which literally just means adding things up to produce tables of numbers, and that produces the official formal numbers. But even then, I don't think that the story of data is done. I think each one of these phases is the data, just in a different phase or different form. The final stage is those printed tables go off into the world and they too, live a very chaotic life, as people choose to use these numbers and interpret them in a wide variety of different ways.
Shawn:There's almost poetic quality to the way that you explain this process in the book, because it could be quite dry or you could imagine, to some people this is all stuff that they're not interested in. They're interested in the end product, and the way that we get to that end product is not of interest to them. I could imagine in certain contexts to some people, depending on what they're interested in their eyes, just gloss over right. But I think you do some really interesting things here, and explaining this in the book is because you could have skipped a lot of that. You could have skipped the process and then just jumped to where I want to go next, which is how we actually talk about people.
Shawn:But before we get there, I do want to say that I think there's a poetic quality to the way that you talk about this in the book, such that when I was reading it actually, like I said at the outset, it felt very dramatic, like I could imagine this happening, the interaction between the numerator and the citizen or the person right and I could imagine the cleaning stage. I just it was something that was very tangible to me. So that's one thing that I'm really impressed with, but the other is that in doing so, by not glossing over it and really focusing on it and parsing out some of this nuance, I think you did an amazing job of explaining why it's so important to pay attention to those parts of the process that might seem like just wrote components necessary to collecting information, that they actually are part of the census in their own weird way, even though we don't see that in the end data.
Dan:Yeah, I mean there's so much. We lose so much when we only look at the final tabular data. But I wouldn't even go that far. Like I looked the way you said it was poetic. I think of it as what I want to do in the book because this is how it works in my life is to be able to look at a table of data and start to have some of the appreciation that, like someone has when they've been taught how to read a poem for the first time, or like how to look at a piece of artwork. I write about this in the book, but it's for real, Like I. Then, whenever I see tabular data I mean a lot of times when I see a table of data, I start just like drawing conclusions or making critiques about its methodology.
Dan:But I am also capable of then also thinking like oh, there's like a lot of sweat and energy and creativity that went into trying to take the world and invent from it the set of numbers.
Dan:And I can learn a lot by trying to figure out like.
Dan:I can learn a lot about the values of the people who made it. I can learn a lot about the way the society is set up that decided that this data needed to be gathered. I can learn a lot about how people with power or the power to count at least see the world, by what they decide to ask and how they decide to then interpret this and turn this into numbers. And then, if I then happen to have access to those intermediary phases, the places where I see, like people's responses and how they get translated, of course, like the person whose job it is to make the tables, their job is not to rap poetic about sorry, to wax poetic about what it is that they're doing. Their job is to clean the data. Someone like me on the outside can then sit there and look at that and say, actually, that person who's doing this work, which can be mind numbing at times, is also its own kind of dramatic activity, as they're charged with taming the chaos and producing from it something that our society needs to make good decisions.
Shawn:So let's follow this line of thought, because I think you're making a compelling argument that if we study only the numbers in end and that in doing so we come to incomplete, maybe inaccurate, conclusions, and this has implications across not just the count and how we talk about the people in the United States, but a lot of research relies on census data across a lot of fields. I've been in a lot of spaces where, in at one point or another, people question whether or not the census might be a good place for us to get some of the data we're looking for. So what are some examples of what we might be missing or getting wrong? And then, what suggestions do you have for researchers and analysts when they do mine this data or use this data to mitigate the damage that might be done by just extrapolating a story or a conclusion from just the raw numbers?
Dan:Yeah. So one thing I would say for I mean maybe the first thing I should say, because for those people who are listening to this and are maybe folks who do spend a lot of time using census data is that actually often the people who use census data have thought about this stuff a lot. So I don't want to feel like you're hearing this and being like, of course I've thought about the fact that this is not precise.
Shawn:We aim to offend here.
Dan:Oh, oops, I didn't read the memo carefully. The people who what do I mean? Again, I think about this is like the people who most know the weaknesses of a data set are the people who constructed it. So this is not the census, but I think about this in my last book, which is about life insurance companies.
Shawn:Also interesting. Believe it or not, you take something that's seemingly dry and it is. You make it interesting. I have read it.
Dan:Wow, all right, I appreciate that there's this. One of the things that happens in that book is I'm looking at how private modes of thinking about insurance are then used as a means of trying to construct the US social security system. And there, early on, congress is holding hearings trying to determine how to set up the system, and they are starting to think about this as having an actuarial basis, which is to say that theoretically there would be this pot of money that's being drawn from people's wages, stored somewhere and invested to gather interest and use to then pay those people some years down the line. That's how it happens in insurance company. You pay premiums, it gathers interest over many, many years, hopefully, hopefully, decades and decades, and then when you die, that money goes to your beneficiaries, or when you retire, it goes to you as an annuity. So as they're doing this, the one of the actuaries shows up into Congress and they are really concerned with how big that reserve fund is going to be, not because they're concerned about having enough money down the line. They're actually really concerned about having too much, too big a reserve fund that then gives the government too much power to invest this money. It's funny to think about the things that were driving their concerns.
Dan:But remember the actuary giving testimony and he says all right, I'm going to give you a number about the size of this reserve, but I need you to know, before I tell you this number, that it's wrong. You, congress, can change it instantly because and you probably are like in 10 minutes you're going to like pass a law which is going to change some procedure about how this is paid or who's can be a beneficiary, and suddenly this number is going to go out the window. But you insist on me giving you a number. So I'm going to give you a number, but don't take it seriously. Having done this, he gives them the number and it's printed on the front on all these newspapers the next day as, like this is the size of the reserve, all of the caveats having been lost.
Dan:And I think this is the fundamental problem that people who produce data sets face, which is that they are often called upon and required by later data users to produce precise numbers, and they are aware of the fact that sometimes there's measurable uncertainty, and then there's all these different forms of unmeasurable uncertainty that lead in and feed into producing their data. And the question, the real question, I think, is so how do you find a way to try to get data users to take seriously those forms of uncertainty? How can you try to break this up? And since Spiro has methods, I think we can probably talk about some of those methods, about how it talks about uncertainty, but I don't think that I'm the best person to give the answer. How you fix that. I think that's identifying. The problem we need to identify is how can we not just communicate uncertainty but build systems that understand the precision that we seem to have with our data is not so precise as it looks like it is.
Shawn:I want to pivot to talking a little bit about United States people and how we count them and how the ways that we count them create their story in a way. So let me start here. You in the book use the phrase statistical margins of society, and I'm not sure why this felt so profound to me, but it really was, and I think that this phrase statistical margins of society captures a lot in just those four words. It captures the idea that society has a hegemony, that some people live on the fringes and that how we count these people both reflects and influences how we think about these people and people in society, and that the narrative that translates from that is that some people are maybe not important, at least at certain points in time. So in your research for the book, I guess I'm wondering how was this showing up, or how were you seeing this or ingesting this?
Dan:So I came up with that phrase. I mean part of it's just evoked by like, literally, you look at these pieces of paper and there's some answers that fit inside, that are allowed. Like there are eight allowable I've just made up that number, I can't have to count like allowable racial categories that can be included, and so you know that there's going to be people who in fact aren't going to find listed a race that's going to properly describe them, and so like there's literally, like sometimes you will see spilling off into the margins of the paper people trying to explain or offer more information, and so, like, when I think of the statistical margins, I literally mean the margins of the paper, but also like we talk about marginalized people or marginalized groups, and one of the reasons we like using that kind of frame is that it makes it clear that somebody was the marginalizer or like society has produced a norm, as you put it, a hegemony that then makes it such that others fall outside of it. One way that this manifested would be like with the homeowner's loan corporation maps. So this is not from the census data, but it's stuff I was using to try to think about who might exist on the statistical margins. So your listeners have probably seen these maps before.
Dan:When they hear the phrase red lining, probably what flashes in their minds is one of these maps in which there are some communities that are literally painted in red on these old 1930s maps of communities. And that then these? We understand that these were used as a means of trying to decide who would get access to credit for mortgaging their houses and who didn't. It turns out it's actually a pretty complicated story about how these maps actually directed people's decisions, so we'll leave that to the side. What we know for sure, or what's easy to say, is that those homeowner's loan corporation maps were a representation of how elite bankers, elite real estate agents in communities saw those communities. So when they colored in an area red or yellow, they were indicating we believe that this area is a higher risk space. It's a space where we think there's either more poor people, there's more people of color, explicitly racialized. We think it's a place that the buildings are of lesser quality. It is a literal judgment and a means that we can figure out.
Dan:Okay, so if I was an elite in the city, what did I think the good neighbors, what's where? What did I think the bad neighborhoods were. This map tells me that, that's the thing it tells me most directly, and so I thought of that like all right, good, this is the thing that shows me. This is what the statistical margins of the society are. I guess the final thing I'll say is the margins can go the other way too. Right you can be.
Dan:This isn't a bad kind of marginalization, maybe, but like one can exist on the margins or outside of the site because of being exceptionally powerful or exceptionally wealthy. So for the 1940 census, like I said, there was an income question and it was meant to record each person's income precisely to the dollar. Unless they made more than $5,000 a year, which at that time was quite a lot of money. Someone who made more than $5,000 would just have it marked down as $5,000 plus. So in that case this isn't a sign that people weren't important. It's a sign that they are really important and in fact so important and powerful that they can manage to make themselves disappear and not have to give precise information about their income to the census taker.
Shawn:So one of the things that I think that in discussing marginalization that we as a society struggle with, and that I struggle with, is threading the needle in that conversation in such a way as to determine what is marginalization by harmful intent and what is marginalization as a byproduct of something benign, benign ignorance, or just a benign byproduct of a behavior that was in no way considering marginalization. The reason I'm mentioning this is because this is something else that I think that came through in the book in focusing on each of the different actors at each different stage and how critical their role was in collecting this data and then producing this data about people in the United States, and that in some way reflects these people that live in the margins of society. That there are certain stages that are simply byproducts of the process that create marginalization or enhance marginalization, and there are certainly some stages and processes that have that intent, and just how complicated it is to have a true conversation about marginalization without considering these two potential influencers, and I'm wondering if you have thoughts about that.
Dan:So I'm led again back to the way we've been talking about our different ways of conceiving of the holes in the data. I guess, as we talk about that distinction in terms of how and why someone is marginalized, or how in my system, produces marginalization, and so maybe this is like. Maybe I will partly dodge and partly answer your question by pointing to the chapter about partners, which is certainly one of my favorite pieces. So early on in my research, somebody asked me. A colleague said so. I found these two people in the census and one of them is labeled head in the household and then the next person with whom they're living is labeled partner. There's two women. I think they're probably two women, harlem although the person who asked me was a little bit cagey about it because they wanted to write about it themselves. But they said what does it mean that these are partners? And I had never run into any partners.
Dan:It was very early on in the project once I started to investigate it and going through their head, my head, everyone's heads, it was like all right, well, we know that these days a partner can mean lots of things, but it often means kind of like an intimate partnership. Is this what was happening, and I think that we see queer communities or queer couples show up in these census records in ways that we wouldn't expect them to show up. Well, I mean, yes, the answer is yes, but partly the answer is yes, and this is where I think I am still getting to. This question you're asking is because we have to take queer history, or looking at kind of queerness, and use a very expansive definition of what it means to be queer as explicitly not just looking at for gay or lesbian couples, but really looking for people who exist outside of what the straight norms are, who form households that don't fit what the straight norm is intended to be a man and a woman with children together living in a nuclear family Anything that's outside of that suddenly counts as being essentially a queer relationship.
Dan:And what I, what could be, you know again, like frustrating in some ways, is that, like we don't know in many cases why a person is listed as partner. There are, like some again fascinating characters full of mysteries here. My favorite two are two women living together in Greenwich Village in Manhattan named Lee Lusgarden and Tessie Finger, and Tessie Finger was, of course, a stenographer, like the perfect name for the perfect name, of course, and Novelis too, made this stuff up. You'd be like that was a little bit too heavy handed.
Shawn:Too brute force yeah.
Dan:But I don't know anything about Lee Lusgarden and Tessie Finger really, so like that is kind of frustrating. But while I don't know anything about their relationship, what I did determine was this is a really interesting queer community here in Greenwich Village. You're like oh thanks, dan, I didn't realize the Greenwich Village was a queer community. Well, this can be expanded to other places too. Like you start to find partners showing up in the sense of manuscript census in lots of different places and part of what you figure from this is like all right, it's showing me not just a place in which people could have gay relationships, but it's also showing us a place where all kinds of other different family formations could exist and often did exist and could be counted. Because my kind of sense is that probably an enumerator in a queer community wandering around pretty soon is disabused, or pretty quickly is disabused of the idea that they're going to find a bunch of readily countable nuclear families in the traditional straight sense and thus, having their eyes opened, need other ways to figure out how to make people count, and so they use this partner category to do this, and so, yeah, it produces a lot of holes. It shows us people who are literally marginalized. It's like the marginalization is not just these people get counted, all the folks living in what we're going to call a queer community broadly but then they are ultimately erased from the numbers.
Dan:So we know they exist now, after the manuscript data is released, but at the moment of the counting, because the Census Bureau didn't see queerness or queer couples as a viable counting option all of those people who had been labeled partners were edited, revised, punched in paper cards as lodgers and showed up, ultimately, at the time, only as categories of the lodgers. So the tragedy there is not a tragedy for us now in which we can see back and start to recreate these strange and complex relationships. The tragedy is to think that this was yet another official statistical manifestation of the erasure of queer life that was happening constantly at this time, and that's a tragedy to understand. It is also, though, an important thing for us, a way that we gain knowledge about the past by thinking about all of the parts of the data. It helps us really understand exactly how it was that queerness both survived, persisted, was honored by the people who were living it, and then was suppressed and pretended to be non-existent by officials with the power to make the final statistics.
Shawn:Well, I'm glad you brought us here because this is a good segue. I don't think this was the intent. This wasn't the main theme of your book, but I noticed in reading it. As a queer person, I'm always interested in uncovering queer history and I think that what you did in a way highlights how difficult it can be for marginalized communities to document their histories and stories because of the ways that we as society choose to count and acknowledge and memorialize non-dominant communities. I think you just outlined very eloquently and you do so in the book by way of example the use of the word partner on the census. But I know, as a queer person I'm really hungry for understanding how the community has evolved, how the community has shown up in history and how that might inform the community now, and I don't think I'm alone in having that kind of hunger or that interest. Do you think this means that for some of us queer folks and folks of color, immigrants, women have just maybe lost critical parts of our stories to history?
Dan:Oh, yes, I mean I don't want to laugh like it's funny this is like the laughter of great tragedy, Like yes, no. The answer is yes. Queer folks, people of color, immigrants, any group that has been pushed to the margins, huge amounts have been lost. I mean that can still be part of the story, right? Like often, then, one of the things we do is we try to explain that story of marginalization, because then that's part of a means of thinking about all that those who came before us endured and the way in which, like you know to be queer today, we get to to trod on paths that were opened up for us by the hard work of the queer people who came before us. At the same time, there are also methods we can use to try to help fill in some of those, some of those gaps.
Dan:Say, D Hartman, scholar of African American history and literature, talks about this and she uses the phrase critical fabulation, and this is a kind of method I felt I was sometimes using as well, which is to say that, faced with records that have been, that have erased people systematically, we're left only to use our imaginations as an act of retrieval and a means of trying, when impossible. You write, you work as hard as you can, and then you we always in history run up to a place where we run out of facts. And then the question is like do we just simply give up, or is it reasonable? And I think we answer many of us has. It is yes, it is reasonable to start speculating precisely when we're dealing with communities who have, when so much work has been done to erase them or to make them make it not visible, to have lost this history. We, we tried to do works of repair, to like find a way towards those stories.
Shawn:So it strikes me that we have come all this way and we haven't mentioned one of the words in your book's title, and that word is democracy. So let's fix that. I think a central tenet of your book is that our American democracy is only as good as the data, but I'm wondering if you could maybe put some flesh on that bone. So how is our democracy informed by census data, and could we make it better if we did things differently, and how we implement and conduct the census?
Dan:The census exists because the United States Constitution built its theory of representation on population. So it said that one branch of government, the House of Representatives, and then the means of electing the president, the electoral college, would be apportioned, would be like, would decide how many, how much power each state had according to its size. And so, because the Constitution had that in it, this whole operation had to exist and we had to start counting people. So that's why it seems reasonable to call this democracy's data, like the entire democratic system of allocating power in this particular federal society is dependent on getting account. Ideally that's an accurate count, but at the very least it has to be a count that everyone's willing to agree on and use. And then, for most of the history of the United States, congress would get from the Census Bureau, or from the Census Bureau didn't exist for most of this time. So we'd get from the Secretary of State, whoever was in charge of this, a count of the number of people living in each state, and then their job was to figure out what do we do with this, and for most of the time, they would pass a new law, and that new law would say when the next Congress starts, this is how many seats there will be in the House of Representatives, and they usually increased it. And then this is how many seats each state will get, and there would be a list. And that apportionment amongst each of the states was meant to be proportional Turns out the math of breaking that down is complicated.
Dan:There's a number of different ways to do this to make it proportional, because you always end up with fractional pieces. They go through a bunch of different techniques. That's not quite so important. One of the things that surprised me in doing this research was I realized after a little while that after 1910 or 1920 on, the size of the House has been constant at 435 seats, and this maybe isn't surprising really, when you just put it like that. But what had been surprising was that before the normal thing was that the House, the size of the House, would increase with the size of the population.
Dan:Ever since it's been frozen in 1920, we've had a way in which this apportionment of representation happens breaking up, taking seats away from states that are shrinking or growing more slowly, giving seats instead to states that are growing more quickly, and that can be like a pretty serious thing to take away a seat or to give a seat to new states and at the same time, that we've tripled the population of the United States. So we have this thing where we have a very high stakes calculation happening how the count of the population matters, and sometimes it can be like a few hundred seats, a few hundred people counted, who decide whether or not a seat goes to one state or another. High stakes decision and it's being based on numbers that we know. The Census Bureau is going to come out with a coverage measurements survey a couple years later. That's going to say, yeah, this is a great count and our uncertainty is in the order of 100 to 200,000 people in every state. So the uncertainty is much larger than the differences we're often talking about in making these decisions about allocating representation.
Dan:So it's like that was for me of the sense that like, oh, this is so crucial to how we set up our democracy, but it shows that because we've frozen the size of the house, we're relying on the data to make decisions for us at a precision that it just doesn't have, and then the final cost is just that we've made it so that we, as each individual person, have so much less representation and so much more distance from our actual Congress people. I didn't know that I thought the House should expand before I started this book, but then, having worked in the book, I suddenly thought oh, this is amazing. This is like very bad that the house has been frozen for so long and we need a lot more representatives.
Shawn:One of the arguments whenever this conversation comes up with some amount of sincerity in Congress, one of the arguments against it is that Congress is already so large that it would just become unwieldy. But that's ignoring the fact that many democracies around the world have much larger legislative bodies that function seemingly at least contemporarily more efficiently. But we're nearing the end of our conversation and I guess I want to ask you a question that you lead with in the book and that is here we could cue like some swelling symphony what is our American democracy if this is its data?
Dan:So I want to answer in two ways. One way I just want to quote my friend, mita Anand, who is kind of like the head of all things data and data equity at the Leadership Council for Human Rights, and Mita has this way where she talks about the census as a report card for the country and what she means. I think in particular she's thinking about the way in which we know that there's a persistent undercount of people, particularly of people of color and also of other kind of marginalized groups, by the census, and part of what she's getting at is that there are ways, there are distinct ways in which we can do things that would decrease the undercount, but part of the best way to get rid of an undercount would be to build a society in which everyone wanted to be counted and was in a stable situation to be counted, because it's not even a lot of it like wanting to be counting it. When we think about who doesn't get counted, it's often those who are have the least resources or who don't have a stable home or who don't have access to many of the systems that make counting something easy to do. Right, you might just be not having high speed internet these days and so, like the census, in a truly equitable society we'd probably have a better count, precisely because people would all live in such a fashion that they would be easy to count and readily countable. And to the degree that this is difficult, it can show us some of the ways in which there are persistent and difficult inequities in the way our society runs. But the other answer I mean, I think there's some hope in that one but then even more hopeful answer is to say when we can look at the system right in the book I talk about some pretty terrible stuff that happens with the census, including the means by which the census data is turned against Americans, american citizens, even during the Japanese incarceration during World War II, but in its like, absolute best form.
Dan:Right, if you think about like, how does the democracy's data tell us something about this place? I think the most fundamental value and one of the reasons I wrote the book in the end, like was there's something really valuable in the idea that every person is supposed to count. It doesn't always happen perfectly, but every person is supposed to matter. They're supposed to then be counted. They were supposed to put a lot of energy and resources into counting them and then we're going to preserve their privacy for 72 years but hold on to their records so that, when everything is said and done, they have a place in history, especially the marginalized and those who about whom there's seldom going to be records kept. They are kept in this place. So there's reasons to not be to look at the data, the census, and say like, oh, america has a long way to go and that's right. There's also reasons to look at and say like there are some fundamentally powerful values that are instantiated in their being a census, and that's pretty cool.
Shawn:All right. Final question Are you ready for it? I'm ready. What's something interesting? You've been reading, watching, listening to or doing lately.
Dan:I mean I can't just say one good book that I was listening to, so I was thinking about this, and there's a few really great ones. One I want to call out Rachel Swarance is a New York Times journalist and she's got this book I would call the 272, the 272, the families who were enslaved and sold to build the American Catholic Church, and she's doing some of the stuff we were just talking about. She is at one level. She's trying to explain why it was that Jesuit leadership running Georgetown decided that they needed to sell enslaved people in a mass sale in the 1830s, and she tells that story really, really well. At the same time, she's also doing the work of trying to reconstruct the lives of an entire enslaved family and then their descendants, with a real emphasis on the extent to which they were fighting for their freedom and asserting their freedom throughout the entire time that they were enslaved and really honoring their own desires and the lives that they were building for themselves. And she just does a great job. She sits on the shoulders of the members of this family and she puts us on the shoulders of the folks who are trying to run Georgetown and really asks us to try to understand both of them. And that's just such an important thing to do at this moment because a lot of us are tied to institutions that are both thinking about how to take care of and actually do some of the reparative work from the long histories of exploitative systems like slavery, and also because we're also in institutions or belong to nations that are right now tied up with and invested in really messy system of exploitation or systems of global climate change that need to be changed, and I think Swarons is getting us to think about that.
Dan:I'll say more briefly the other ones. People should read Ellen Alman's Close to the Machine. It's just my favorite memoir from a tech writer. It's now 20 years old, but she really tries to again help people to understand why it is that allure of being the computer programmer is, and especially the allure of using that as a way to escape from the messiness of humanity. And then two books that are seemingly about thinking, but in very different ways. Death Pop Berman wrote the book Thinking Like an Economist, which tries to explain why it is that economists seem to do so much public policy these days, and it's just brilliant. It's really good, she's right. And then the other one is by historian Jamie Kreiner. It's called the Wandering Mind, what medieval monks tell us about distraction, and it's just a hoot and she really gets us to think about how long and then how many different ways people have been trying to understand and think about what it means to think and what it means to be distracted. This is not like a new problem that we face today.
Shawn:Typically, I've heard of one book someone mentions and I'm always looking for books to read. You just mentioned four that I have to add to my list and I haven't heard of these, but you make them all sound so fascinating, I have to say related to the 272, so there's a bit of history, the Catholic Church here and I have to say the churches really have in a moment right now and it ain't all good right.
Dan:No. So Warren's is interesting Again, just like she's such a brilliant book that she constructs, because part of what she's doing then. The book, though, is also later on explaining how and why for formerly enslaved people, it was really important to be able to hold on to their Catholic faith, and how they dissociated that from the reality of the means by which the power structure had enslaved them. So I mean, as you probably picked up from, and listeners will pick up from this like I'm all about my ambiguities, and she just does it really well.
Shawn:Dr Bouk, it's truly been a pleasure. Thanks for taking the time to stop by this is wonderful, thanks for having me.
Shawn:Census data and its method of collection are the linchpins of a thriving democracy At its core. Census data represents more than just numbers. It reflects the heartbeat of a nation, the diversity of its people and the power of their voices. It's not just a count. It's a vital reflection of our society, revealing the diverse tapestry of voices that come together to form this great nation. As Dr Bauch has highlighted both in his book Democracies Data and again today in our conversation, how this data is collected directly impacts the very essence of democracy and its core ideals. And if we see census data through that lens, then it isn't merely a snapshot of a number of people in a specific place on a certain day. It's a compass guiding us toward a more equitable future. It informs policies that touch every aspect of our lives, from education and healthcare to infrastructure and social programs. A complete and accurate census empowers policymakers to make informed decisions, ensuring that no community is left behind and every voice is considered.
Shawn:The challenges we face in achieving an accurate and inclusive census count are substantial. Overcoming historical barriers, fostering trust and embracing modern methodologies are crucial steps toward enhancing the accuracy and completeness of this essential data set. By addressing these challenges head on, we pave the way for a more representative democracy, one that respects and acknowledges the rich diversity of its populace. So, keeping this in mind, let's carry forward the understanding that the census is a reflection of who we are as a nation and a testament to our commitment to democratic ideals. As Dr Bauch suggested, let's engage, educate and advocate for a more inclusive, accurate and representative census, because it's through this shared conviction that we can build a stronger, more united America. Alright, a final update I will be out of country for the next few weeks, so deep dive will be on a short hiatus. In the meantime, though, you won't be left hanging each Sunday while I'm away. A deep dive greatest hit will release, so keep coming back. Chat soon, folks.