Religion in the United States seems to be understood by all Americans. News sources talk about keeping religion out of politics, the Supreme Court debate cases regarding the separation of church and state, and the First Amendment explicitly guarantees every citizen’s right to exercise their religion. Very rarely though, do Americans sit down and really consider what is even meant by ‘religion’. In fact, considering definitional boundaries and their consequences is a practice most Americans could use brushing up on.
This is because information is becoming more and more accessible. With this influx of information comes added responsibility for the knowledge consumer. No longer is knowledge as challenging to obtain. Libraries are transitioning books out and computers in. The days of phoning a friend and actually speaking to them instead of texting seem far away. Touchscreens have made information available at our fingertips, but for all the modern ease of accessing information comes the added difficulty of deciphering that information.
More information means less accuracy and the fast-pace of producing and sharing knowledge through technology opens more doors for mistakes. Everything from simple typos to blatantly falsified information — in this day and age, everyone gets “published” whether they might be misinforming the public or not. As explained in Digital_Humanities:
“the mere existence of vast quantities of data, artifacts, or products is no guarantee of impact or quality”(Burdick, Drucker, Lunenfeld, Presner, and Schnapp 2012, 34).
The vast quantity of data online does not equate to reliable, thoroughly researched information. But more often than not, complex nuggets of knowledge are often presented as straightforward and simply factual without regard to the many sides that accompany any narrative.
This modern idea of progress alongside regression has been discussed innumerable times with innumerable arguments. The issue is not that Americans are losing intelligence or are incapable of thinking critically, but that the fast pace of sharing knowledge has left little room for the necessary in-depth evaluation of a source. While everyone should certainly take a moment to reconsider before retweeting or pressing “post”, the heart of this problem is one of definition and is not a new concept.
Modern conversations about technology make it seem that sharing one-sided information is a new issue. As if it is a new problem that Americans must cope with as technology advances and more humans are virtually connected. But, long before computers and social media existed, people roamed the earth with just as many social issues as Americans face now. The problem is not one of internet vs no internet, but of evaluating the implications of the definitions on which social actors rely.
This problem of thinking critically about how social actors define their environments is not a new problem, as I have pointed out, and I hope to demonstrate the power that individuals yield when choosing how to define a particular concept. Whether it be a published paper from a scholar or a tweet from the President of the United States, every piece of knowledge has a source, and re-evaluating that source is key to achieving a better understanding of the situation or concept as a whole.
One of the easiest ways to show the issue of definition is with a visualization of data on religious families in the United States. I created each visualization below from public datasets that all share a similar goal: cataloging self-identified religious families in the United States at particular moments in time. The first, is from The Association of Religion Data Archives (the ARDA) in 2010, followed by a survey from the Baylor Religion Survey in 2010 and 2014 and another survey conducted by PEW Research Center, also in 2014. I will describe each before analyzing their relationships with one another and evaluating what those relationships might imply. From there I will demonstrate how the real-world implications of knowledge production will require dissent from tradition for scholars in the field of religious studies and Americans more generally. More information about the collection and visualization of these datasets will follow in this project’s Process Statement.
ARDA Data 2010
Baylor Data 2010
Baylor Data 2014
PEW Data 2014
How many religious families existed in the U.S. between 2010 and 2014?
This question is difficult to answer. After evaluating only four different datasets, it is obvious that finding consensus on the number of religious families is a bit harder than it first appears.
The dataset from the Association of Religion Data Archives uses 6 different categories to distinguish religious families from one another.
The dataset from Baylor University in 2010 uses 41 different categories to distinguish religious families from one another.
The dataset from Baylor University in 2014 uses 41 different categories to distinguish religious families from one another.
The dataset from the Pew Research Center in 2014 uses 52 different categories to distinguish religious families from one another.
This means that in 2010 two different surveys were conducted by two different research parties. One survey identified 6 religious families and the other identified 41. Between 2010 and 2014, the same research survey used at Baylor University maintained 41 categories. But, in the same year, a different survey from the Pew Research Center identified 52 religious families. Three different sources focusing on two different years identified between 6 and 52 different religious categories.
How are Religious Families between 2010 and 2014 being defined?
This massive discrepancy in the number of religious families is a result of the choices made by the surveyors based on their unique goals. The oftentimes misleading understanding of what it means to observe is expanded on further in the Process Statement. For now, I will describe how each survey was designed with a particular goal in mind based on the way they categorize religious families in the United States.
The largest categorical grouping from the Pew Research Center is plentiful because it is broken down into several subcategories. The ARDA dataset is the most restrictive, with only Black Protestant, Evangelical Protestant, Mainline Protestant, Catholic, Orthodox and Other as religious family options. On the other hand, the data from the Pew Research center — the largest categorical grouping of the examples — is plentiful because it breaks down Protestantism into varying subcategories.
While the ARDA keeps its data groupings generalized, the Pew data is highly specified and almost flips the organization. Where the ARDA has simply “Black Protestant”, the Pew data has categories like “Methodist in the Historically Black Protestant Tradition”, “Nondenominational in the Historically Black Protestant Tradition” and “Pentecostal in the Historically Black Protestant Tradition”.
The data from Baylor University, like the Pew Research Center data, is also more specific than the ARDA data though it has fewer subcategories than the Pew data. The Baylor data has no mention of groups like Black, Evangelical, or Mainline Protestant. Instead, the categories seem to find a middle ground between the broadness of the ARDA data and the specificity of the Pew data. Categories include “African Methodist”, “Nondenominational Christian” and “Pentecostal”.
None of these groupings are right or wrong, but they all serve as an excellent example of the different ways conclusions are drawn about the same information. Technically, all of these sources evaluate religious families. But the conclusions that are drawn from each of these vary greatly when evaluating the data only as it is presented (in other words, without combining Pew’s “Nondenominational in the [Historically Black/Evangelical/Mainline] Protestant Tradition” into one grouping titled “Nondenominational Christian” as was done with the Baylor data, or further into simply “other” as is done with the ARDA data). In fact, the lack of similar naming of each religious family makes it difficult to compare across data sources.
Even the groups that are named the same show discrepancies. Let’s briefly look at the category of “Catholic” across all the sources. The ARDA data shows that about 35% of the individuals surveyed are Catholic in 2010 but the Baylor data shows that number as about 22%. In 2014, the Baylor data shows 27% of those surveyed identified as Catholic and the Pew Research data shows 20%. This comparison is not made to try and falsify the fact that 20% of the individuals surveyed by Pew self-identified as Catholic, but that viewers of that specifically collected data have a little more work to do before they can draw wholesome conclusions about what those numbers might mean.
This extra work varies on the knowledge consumers needs, just as the producers of each survey formulated their categories based on the goals of their particular studies. Religious Studies scholar Russell McCutcheon points out that we need categories to organize our world:
“[possibly arbitrary] standards need to be established and invoked for any judgement to be made about anything”McCutcheon 2007, 19
So the argument is not about whether the category “other” in the ARDA data is the same as “No religion” in the Baylor data or the multitude of other/none options in the Pew data, but that something is being accomplished in categorizing certain groups as religious families or not religious families. McCutcheon expands:
“Classification is therefore a social act that helps us to make values, establish relationships (often they’re ranked relationships, since clean is usually assumed to be better than dirty), and thereby manage the world.”McCutcheon 2007, 19
While classifications are necessary to organize the world, they are never unbiased. That is not to say that classifications are bad and should be avoided, but that they should be handled with care and deliberation when consuming and reproducing knowledge.
What is gained in being named a religious family?
The way we define terminology matters for many reasons. For one, consider the financial implications of being considered a religious family or organization. As described on the First Amendment Encyclopedia webpage:
“Today, in all 50 states and the District of Columbia, statutes and constitutions provide various types of property tax exemptions for religious organizations. Similarly, the federal government has exempted churches and other religious organizations from federal taxation in the modern federal tax code since the ratification of the Sixteenth Amendment to the U.S. Constitution in 1913.”Rosenblum 2009
Naturally, several court cases were born from these exemptions. Most of which evaluated the role of church and state and how funding might interfere with separating the two. Eventually, this led to the Lemon V. Kurtzman case in 1971 which resulted in the Lemon Test. This test is used to determine if government involvment in religious matters meet the standards set in the Establishment Clause. It outlines:
- The statute must have a secular legislative purpose. (Also known as the Purpose Prong)
- The principal or primary effect of the statute must neither advance nor inhibit religion. (Also known as the Effect Prong)
- The statute must not result in an “excessive government entanglement” with religion. (Also known as the Entanglement Prong)
Defining who counts as a religious family has been outlined across U.S. history. From the ratification of the Sixteenth Amendment in 1913 to the Lemon case in 1971 to President Trump’s photo-op in from of St. John’s Church this June. Being chosen to represent a capital R “Religion” matters financially, socially, and politically in the United States.
Consider the fictional adult animation show Bobs Burgers that follows a family in the Northeast U.S. who owns a family-run burger shop. In 2017, an episode called “Aquaticism” was released on FOX. The episode opens with Bob’s three children (Louise, Gene, and Tina) at local aquarium that is on the verge of closing. After the mortician next door explains tax exemptions for churches, the Belcher kid’s are inspired to keep their financially failing, local attraction afloat.
And so, a new religion is born: Aquaticism. Things take a turn when the IRS investigator decides he wants to convert to the local religion. He believes initiation into the church requires jumping into a aquarium of live jellyfish, and the kids are forced to confess their lie before he is seriously injured. For this fictional family, the classification of their favorite aquarium (or keychain store) as a religion means much more than a simple re-labeling.
Although a silly example from a fictional show, this satirical episode serves as a good example of the way that the words, classifications, and definitions that American’s use and thus reinforce matter for real people on the ground. There is always something at stake when naming is involved.
To get back on academic ground, consider Russell McCutcheon’s opening chapter in Studying Religion: An Introduction. Similar to my efforts here, McCutcheon complicates widely accepted names like Mount Everest to show how the names given to objects matter. He explains:
“The same generic material takes on different meanings, values and identities in relation to different classification and thus organizational systems (just as that large snowy mass did when the surveyors arrived), each of which puts into practice different sets of interests—which changes from time to time, group to group, and occasion to occasion.”McCutcheon 2007, 19
The point of this project then is not to decide which source most accurately described religious families in the U.S. between 2010 and 2014, but what conclusions each survey was allowed to draw because of the way they represented their data.
Why does this matter?
Understanding why and how an object is named, defined, and then used is key for responsible scholarship and invaluable to the general public in this day and age of rapidly growing access to information. Lateral research, evaluation of sources, and just general consideration of what is accomplished when data is presented without consideration for other possibilities would help anyone better engage with their changing social environments. Scholars especially have a duty to acknowledge their role in knowledge production. Although there is much-earned respect and authority for being a specialized scholar, that authorization should not be used as a clutch to make claims without considering the variants of an argument. As described in Data Feminism:
“When approaching any new source of knowledge, whether it be a dataset or dinner menu (or a dataset of dinner menus), it’s essential to ask questions about the social, cultural, historical, institutional, and material conditions under which that knowledge was produced, as well as about the identities of the people who created it. Rather than seeing knowledge artifacts, like datasets, as raw input that can be simply fed into a statistical analysis or data visualization, a feminist approach insists on connecting data back to the context in which they were produced . . . Along this line of thinking, it becomes the responsibility of the person evaluating that knowledge, or building upon it, to ensure that its “situatedness” is taken into account.”D’Ignazio and Klein 2020
While situating data might seem like an unnecessary step for the everyday individual (and, unfortunately, for many scholars), this form of critical thinking opens the door to a more wholesome understanding. It should be comforting to know there is never one right answer. The world moves too fast and changes even faster to stick to one understanding of how things work. Understanding the implications of definition allows actual social and political action to not only be taken but carried out efficiently. If the Belcher kid’s from Bob’s Burgers can try to save their little aquarium with a switch in definition, just think what has been and can be accomplished in understanding and employing definitions with their multiplicity in mind.
To read more about how I collected and organized my own data, follow along to my Process Statement.