Debating Empirical Questions on the Internet: Rival Claims About Crowd Sizes of the Glenn Beck/Stewart-Colbert Rallies

by Whitney D. Gunter and Joel Best
Western Michigan University; University of Delaware

Sociological Research Online, 17 (2) 6
<http://www.socresonline.org.uk/17/2/6.html>
10.5153/sro.2545

Received: 19 Jul 2011 Accepted: 4 Jan 2012 Published: 31 May 2012

Abstract

Analysts argue that the Internet can democratize the construction of social problems and reduce claimsmakers' dependence on coverage in traditional media. This paper examines Internet claims about the relative sizes of 2010 rallies on the Washington, D.C. National Mall hosted by Glenn Beck and Jon Stewart/Stephen Colbert. Because crowd size is understood to be an index for the popularity of social causes, and because the two rallies were taken to stand for opposing positions in the culture war, numerous advocates offered competing analyses of the crowds' relative sizes. Analysis of these claims suggests that the Internet offers a forum where a host of arguments and evidence can be presented in a short period of time. Yet, although crowd size is an empirical issue, there was little effort to reconcile competing arguments and arrive at general agreement, and, without media coverage, these claims failed to attract much public attention.

Keywords: Crowd Size, Glenn Beck, Jon Stewart, Stephen Colbert, Social Problems, the Internet

Introduction

1.1 The media play a central role in constructionist analyses of social problems (Spector and Kitsuse 1977; Gamson and Modigliani 1989; Hilgartner and Bosk 1988; Best 2008; Monahan 2010). Claims that some troubling condition should arouse concern are ubiquitous, but they must compete in the social problems marketplace for the attention of the public and policymakers. Hilgartner and Bosk (1988) conceptualize this marketplace as organized around arenas, each with a limited carrying capacity. Only a few claims can command attention in any arena, and some claimsmakers are advantaged in this competition. Claimsmakers have very different resources for promoting claims: activists in well-established social movement organizations, experts who can bring special knowledge to bear, powerful officials, celebrated individuals, and media figures all have ties to the media and find it relatively easier to attract media attention to their claims. In turn, media coverage increases the chances that the public will be exposed – and respond – to claims, and that policymakers will devote attention to these issues. In contrast, the claims of ordinary individuals who lack special organizational ties, expertise, power, or status are less likely to be heard.

1.2 At first glance, the Internet seems to have the capacity to profoundly alter – to democratize the social problems marketplace – by giving ordinary individuals a forum from which they can make claims. Cyber-optimists argue that the Internet will have a leveling effect (e.g. Clinton 2001; Rodman 2003; Ito et al. 2008; Williams and Delli Carpini 2004). By establishing blogs or participating in discussion threads, social media, and other online venues, anyone with an Internet connection and sufficient savvy can make claims that are theoretically available to a global audience. The Internet thus becomes an arena with a theoretically infinite carrying capacity, a place where anyone might make claims and shape the social problems process.

1.3 Scholars have been slow to analyze the Internet's effect on the construction of social problems. Analysts of social movements have noted that the Internet provides important tools for allowing individuals to share unpopular views (Daniels 2009); organizing protests (Fisher et al. 2005; Langman 2005), and maintaining commitment among movement constituents (Eaton 2010). But giving new resources to social movements reinforces the traditional claimsmaking pathway in which activism is intended to prompt media coverage. To study whether the Internet is democratizing the social problems process, it is necessary to examine the influence of individuals' claimsmaking. Thus, Maratea (2008) argues that, while blogging is a new forum for claims, bloggers' influence depends upon having their claims relayed by traditional print and electronic media to a larger public.

1.4 This paper examines an instance in which rival claims proliferated on the Internet. At issue were the relative sizes of two quasi-political demonstrations in Washington, D.C. Disputes over crowd sizes have a long history, and the Internet creates new opportunities for individuals to weigh in on this topic. However, these claims ultimately had little effect, suggesting some limits of online claimsmaking.

Washington Crowd Sizes as a Subject for Claimsmaking

2.1 The ability to mobilize a substantial crowd is widely considered to be an indicator of a social movement's strength. A large crowd suggests that a movement has extensive support, while a small turnout implies that a cause lacks broad appeal. When a movement's cause is contentious or controversial, there may be competing estimates for crowd size, with supporters offering bigger estimates, and opponents promoting smaller figures. As in the case of other social statistics, the numbers assigned to crowds are often treated as facts; popular discourse tends to ignore the processes by which estimates of crowd size are socially constructed (Best 2001).

2.2 These dynamics have often been evident for demonstrations on the National Mall in Washington, DC. The National Mall is generally understood to extend from the Lincoln Memorial's steps in the west, to the west side of the Capitol in the east.^[1] It has been the setting for numerous demonstrations, most notably the 1963 civil rights March on Washington, but also countless peace, anti-war, and support-the-troops demonstrations, pro-choice and pro-life rallies, and events such as the Million Man March, major gay rights and Promise Keepers rallies, and so on. In turn, many of these events have inspired disputes over crowd size (Barber 2002; Gitlin 1980; McPhail and McCarthy 2004). For a time, the National Park Service Park Police (the law-enforcement agency responsible for the National Mall) provided official estimates, until Congress, following the rancorous dispute over crowd size at the Million Man March, forbade the practice (Smith and Melillo 1996). With the exception of the presidential inauguration in 2009, the Park Service has stopped releasing crowd estimates (Moore 2009). Thus, there is no longer an authoritative source for these numbers.

2.3 At the same time, crowd sizes remain subjects of interest and dispute, even as the Internet has emerged as a new arena for debating these issues. As recently as 1995, when rival estimates over the size of the Million Man March inspired intense argument, claims and counterclaims were largely restricted to traditional print and electronic media. However, the broad accessibility of the Internet now makes it possible for individuals to weigh in on these debates, to present their own evidence, offer their own estimates for crowd sizes, and criticize one another's claims. The nature of statistical claims on the Internet is illustrated by the online dispute over the relative crowd sizes at two 2010 demonstrations organized by television personalities. Examining this case offers insights about the Internet as a forum for debating empirical claims related to social problems.

Estimating the Size of the Beck and Stewart/Colbert Rallies

3.1 On August 28, 2010, Fox News commentator Glenn Beck hosted the "Restoring Honor" rally at the Lincoln Memorial. According to C-SPAN's (2010a) description of the event, the rally's speakers focused on "the conservative agenda, religious values, civil rights issues, the economy, and reducing the size and influence of government." Initial media estimates for the crowd's size varied dramatically, and included estimates as vaguely low as the Associated Press's "tens of thousands" (Elliott and Syeed 2010), to Glenn Beck's own, larger estimate of "a minimum of 500,000 people" (Montopoli 2010a).

3.2 Anticipating such discrepancies in estimates, CBS News commissioned AirPhotosLive.com to photograph the event and use the images to calculate the crowd's size. Their methodology involved taking pictures of the event with several goals in mind, including getting photos of different areas from different angles, and also focusing on the peak time for crowd size. In the case for the "Restoring Honor" rally, the photographs were taken at noon, which was just after the event's 11:30 a.m. midpoint (Montopoli 2010b). Afterwards, the photos were examined using gridlines to separate the crowd into small units. Individuals in each unit from different sampled areas were counted to estimate crowd density for the unit, and by extrapolation the size of the entire crowd. In order to avoid a miscount in units or having a less than ideal sample affect the estimate, AirPhotosLive.com had three researchers separately use their own methodologies to sample and count the crowd (CNN 2010). They also consulted with Professor Stephen Doig of Arizona State University to provide an alternative estimate (Montopoli 2010b). Ultimately, AirPhotosLive.com's internal estimate was that the crowd contained roughly 87,000 people, while Doig's estimate was that it contained 80,000 people. Both estimates were statistically the same due to a 9,000-person margin of error, and CBS News chose to use the higher estimate of 87,000 (Montopoli 2010b; Sundby 2010). Other news outlets did not attempt to measure the crowd, although NBC's observers suggested that the crowd might have numbered 300,000 (Zernike and Hulse 2010).

3.3 The second rally was the October 30, 2010 "Rally to Restore Sanity and/or Fear" event hosted by Comedy Central's self-described "fake news" personalities, Jon Stewart and Stephen Colbert. C-SPAN (2010b) characterized the event as a "nonpartisan rally" that was a "politically themed entertainment program" and said that the event concluded with "serious remarks criticizing partisan political and media culture that often drowns out respectful debate in the country." However, both Stewart and Colbert had frequently lampooned Beck's broadcasts, and their rally was easily understood as a critical response to Beck's event. Most media sources chose to describe the Steward/Colbert crowd only as numbering in the "tens of thousands" (e.g. Stanglin and Durando 2010). However, as with the Beck rally, CBS News commissioned AirPhotosLive.com to provide a crowd estimate. Using the same methodology used for the "Restoring Honor" rally, AirPhotosLive.com reported that 215,000 people, plus/minus 10%, were in attendance at the "Rally to Restore Sanity and/or Fear" (Montopoli 2010c). Perhaps in keeping with their rally's theme of avoiding unnecessarily contentious rhetoric, Stewart and Colbert did not offer their own – or comment on others' – serious estimates for the size of their turnout, although both made satirically exaggerated claims (Stewart's was ten million and Colbert's was six billion).

3.4 Though not technically an estimation of crowd size, it is noteworthy that the Washington Metropolitan Area Transit Authority (2010) reported selling 825,437 trips for the Metrorail on October 30 (an all-time high for Saturday Metrorail sales), which is about 237,718 round-trips more than usual for a Saturday. In comparison, the Metrorail sold 510,000 trips on the day of the "Restoring Honor" rally – about 80,000 round-trips more than usual for a Saturday (Bolden, 2010). Thus, the two days' above-average ticket sales were quite consistent with the respective AirPhotosLive.com estimates.

3.5 In sum, in the absence of official estimates from the National Park Service, some news media simply indicated that the rallies drew large – "tens of thousands" – crowds; however, CBS solicited estimates from a team of experts at AirPhotosLive.com who calculated that the Beck rally attracted fewer than 100,000 participants, while the Stewart/Colbert rally drew more than 200,000, and evidence from Metrorail ticket sales mirrored those numbers. However, these apparently consistent results did not deter further debate.

3.6 In the days immediately following the "Restoring Honor" rally, Beck and his supporters criticized the CBS estimate that their rally drew 87,000 people. Beck proclaimed that the number was too low and that there was actually "a minimum of 500,000 people," and added, "we'll never know the exact number, but anyone with eyes can see this number was in the hundreds of thousands" (Montopoli 2010a). The highest estimate for Beck's rally came from Representative Michele Bachmann (R-MN), who said that "unofficially, off the record, we talked to one of the guys from the National Park Police who told us he thought it was 1.6 million; there had to be over one million people there" (Diaz 2010). Although the National Park Service no longer makes crowd size estimates, people continue to invoke their authority. Bachmann was not the only one to do so. Domenico Montanaro from NBC News said via Twitter, "There won't be official #s. Parks Svc/police don't do that anymore. But official at top of memorial said 300-325K" (Montanaro 2010).^[2] Citing numbers by an unnamed official who quite possibly has no training in estimating crowds sizes makes non-scientific ballpark figures appear valid on face value as they come from a person of apparent authority, yet the unknown source and methodology is unavailable for criticism.

3.7 The "Restoring Honor" rally's supporters continued both to offer their own larger estimates, and to challenge those offering lower figures. Glenn Beck, for example, said that "the media can diminish the crowd size all they want, but the images speak for themselves…," thus implying that the lower estimates reflected the mainstream news media's bias against him and his agenda (Montopoli 2010a). Because Beck and other Fox News commentators often criticize the mainstream media for being ideologically biased, it was easy for them to challenge the CBS figures. However, it was not until the "Rally to Restore Sanity and/or Fear" that there was a direct rivalry to fuel the debate. With the Stewart/Colbert rally designed at least partially in response to Beck's event, supporters on all sides were ready to use the Internet to make comparisons–and to dispute one another's claims.

The Internet as an Arena for Disputing Social Issues

4.1 Weighing in on the relative sizes of the Beck and Stewart/Colbert rallies offered an excellent opportunity for would-be claimsmakers for several reasons. First, there were no authoritative, generally accepted estimates for the crowds' sizes. The National Park Service no longer issued estimates, so there was no "official" number that media felt obligated to report. Although the AirPhotosLive.com estimates represented experts' judgments, they had been commissioned by CBS News, which may have made other mainstream news organizations less likely to repeat those figures (and, in the process, publicize a rival), and which could be characterized by Beck and other Fox commentators as a biased source. Second, it was easy to understand the Beck and Stewart/Colbert rallies as rival events, reflections of a larger culture war between Fox News's older, more traditional, more conservative audience, and Comedy Central's younger, more irreverent, more liberal viewers. The fact that Stewart and Colbert had frequently made fun of Beck – and that their rally was obviously intended to ridicule his – encouraged people to view the two events in juxtaposition, to take sides, and therefore to favor crowd estimates that supported their positions. Two rival status groups – each invoking a widely held value (honor, sanity) – staged rallies on the National Mall, only two months apart. Which could claim the broadest support? Which crowd was bigger? These questions were of interest, especially to individuals aligned with either side in the broader culture war.

4.2 Third, the Internet allowed participants in this debate the space they needed, not just to make claims, but to make their cases. Through the use of blogs, partisan news sites, message boards, and comments on mainstream news sites, individuals were able to post not just their words, but also images as evidence to support their arguments about relative crowd sizes. In this debate, visual evidence – in particular photographs and maps – could be presented to support different positions. We examine these claims, in order to understand evidentiary rhetoric in the Internet arena.

Methodological Note

5.1 We gathered images related to the sizes of the two rallies using a combination of searches. These included Google keyword searches, Google image searches, and an examination of the comments from users in response to other images and articles about crowd size in which some provided links to other images. For example, a search for 'Beck Stewart Rally Comparison' might yield relevant results in both a standard Google "Everything" search and a Google Image search. Sites would be examined in an effort to find both the original/earliest posting of the image and any discussions about the images. In some cases, these comments might lead to more images as users posted supportive or critical comments and provided a link to bolster their claims. Because images are easy to re-post elsewhere online, the number of unique/original images is low despite the images being used on many websites and forums. We have made every effort to include an exhaustive collection of images, but there is no guarantee that ours is a comprehensive collection. Moreover, we lack complete information on many of the images; some of them were posted via anonymous image hosting sites, or on a forum/blog under a pseudonym, leaving the authors/creators unknown.

Making Visual Claims about Crowd Size on the Internet

6.1 Consider the display posted by "Keypro" (2010) to Free Republic, a conservative Internet forum, at 4:24PM EDT the day of the Stewart/Colbert rally (only about four hours after the rally began). The display (Figure 1) features three adjoining images: the top of the display has side-by-side photos of each event's crowd; while the bottom of the display features a map with ovals denoting the areas covered by the two crowds. The two photos appear to have been taken from the Washington Monument. Because the monument is much closer to the Lincoln Memorial (where Beck stood) than to the Capitol (the Stewart/Colbert stage was located near the Capitol), the former shot appears foreshortened, while the latter crowd appears in the distance.^[3] As the photos were taken at different angles, it is difficult to establish a direct comparison using these images alone. Keypro, however, uses them to do just that, as the crowd occupies a larger proportion of the photo of the Beck rally. To further this point, the third image at the bottom of the display shows the National Mall, with each rally's location indicated with an oval. The Beck rally has an oval that is twice as long and twice as high – that is, four times larger – than the oval for the Stewart/Colbert rally. The image was quickly incorporated into an article (Baker 2010) on The Blaze, a conservative news and opinion website founded by Glenn Beck himself. The article referred to the image as "a photo study."

Figure 1. Keypro’s (2010) display combined three images: (1) photo of the Beck rally (upper left); (2) photo of the Stewart/Colbert rally (upper right); and (3) map of the National Mall indicating areas covered by the two rally crowd

6.2 Shortly after Keypro's (2010) image began circulating, comments started appearing in the same forum thread that either supported the image's message, or criticized its validity. Some critics charged Keypro with using misleading images, while others made accusations of outright lying. For example, NotFoolingMeForAMinute wrote in Keypro's thread:

What this clearly demonstrates to me is that photographic fakery has become quite easy in the Age of Photoshop…. [In CBS's photos of the Stewart/Colbert rally] ^[4] the crowd goes back more than twice as far as is depicted in this obviously faked photo and spills out into the outlying areas…. Now, let's all start proving Jon Stewart right [about insanity prevailing] and let the namecalling begin. (NotFoolingMeForAMinute, as posted to Keypro 2010)^[5]

6.3 Other users voiced support for Keypro, or even said his image was unfair to Beck's rally:

The comparison pictures shows the Capitol 5 times larger than the Lincoln Memorial, thus making the crowd in the 10-30 gathering look at least 40% bigger if you are going to compare pictorially the picture of the Stewart crowd needs to be shrunk 40% or the Beck picture needs to be enlargef 40%. Using Google Earth and the Google Ruler. Thanks Google Earth. Beck Rules. God I love America! (Seven, as posted to Baker 2010)^[6]

Rebuttals to the image also appeared elsewhere on the Internet. For example, a blogger known as "Catholicgauze" (2010) posted an edited version of Keypro's image (Figure 2):

Figure 2. Catholicgauze’s (2010) corrections for Keypro’s map.

Let us examine the geography and image interpretation used for the [Keypro] image. For Beck's rally look how the oval starts to the left of the Lincoln Memorial, where the speaker's were and roads prevented people from being behind, and goes away past the World War II Memorial at the end of the reflecting pool, where the image ends. Now for Stewart's rally look how the at the right side of the photo, there is an oval building. The bulk of the rally ends at the street coming right after the circular building.. The building is the Hirshhorn Museum yet the oval in the image for Stewart's rally ends well before the museum.
Take a look at the mini-map below now with my modifications (a red block for Beck and yellow for Stewart). Size-wise Beck's rally is slightly bigger but Beck's rally had a reflecting pool in the middle of it.
This is not meant to be a political post. I do not even like getting involved in the beast known as population/crowd estimates. This post merely is to show how a map is being used to distort a current event.

6.4 It is worth noting that Catholicgaze's blog post – criticizing Keypro – was dated 8:48PM EDT October 30, 2010 – about nine hours after the rally began and four hours after Keypro's post.

Figure 3. Map by Anonymous (2010a).

6.5 Successive contributions to the debate added refinements. A similar image (Figure 3), created using a map of the area with crowd areas shaded, also circulated the Internet forums (Anonymous, 2010a). This version added greater detail, with areas off limits to attendees (e.g., reflecting pools, buildings) left unshaded so as not to exaggerate the crowds' sizes. Though no sites could be located that pointed to an original source or time for this image, links to it (e.g. Under the Mountain Bunker, 2010) appeared as early as October 31, the day after the rally. A second, widely-circulated image (Figure 4) also uses the shaded area approach (Anonymous, 2010b). It is similar to the previous attempt to exclude areas off limits to pedestrians, but with some subtle differences (it excludes the areas roped off to the public for grass regrowth, whereas the other images do not). This image also references the AirPhotosLive.com crowd estimates for both rallies. As before, an original source/time could not be located, but links to this image also appeared as early as October 31 (as linked by Reverentjester, as posted to Baker 2010). As a side note, it is interesting that the creators of each map chose red to denote the "Restoring Honor" rally and blue for the "Restoring Sanity and/or Fear" rally, adopting the standard ideological color-coding to characterize the widely-presumed political orientations of the crowds.

Figure 4. Image by Anonymous (2010b)

6.6 As with the previous images, online comments continued to debate the validity of the images. For example, one commenter wrote of the second anonymous image (Figure 4):

Thats Bull Crapp! I was at both Rallys and I can tell you that Glenn Becks Rally FAR OUT NUMBERED Colbert and Stewarts Rally. Beck had people from the washington Monument to the Lincolin Memorial and on both side of the reflecting pool and in the fields beyond the trees all the way into the streets,Colbert & Stewart only had half way between the Washington Monument and the Lincolin Memorial.The huge field was empty! Why can't you tell the correct TRUTH? (Rebecca56, as posted in Double Dutch Politics, 2010)

To which user RyanC1384 replied, "Considering you were at both rallies then you would know that the stages were in different areas??" (as posted to Double Dutch Politics, 2010).

6.7 When this image (Figure 4) was posted to reddit (2010), it reached a score of 2,437 (upvotes minus downvotes) and received 1,142 comments. The commentators at reddit likely favored the image's implication that the sanity rally was better attended, and generally made comments about the image that were favorable. For example, Paranoid_Freak said, "There were still 87,000 people at Beck's rally, and that's kind of sad." The most popular comment, posted by Wrym, stated, "Numbers aside, Beck's crowd was much denser," likely implying that crowd at the Stewart/Colbert rally was more intelligent than the crowd at the Beck rally. This comment may have benefited in popularity from the combination of humor and the side of the debate it supports. Conversely, the least popular response, by Rosebowl23, said, "Jesus Christ, now liberals are inventing their own version of reality. Oh wait, they've always done this. Nothing new here" (postings to reddit 2010), thus further emphasizing the site's users' political preference in upvoting and downvoting responses.

Figure 5. Maps using estimated crowd densities by the Lame Texpatriot (2010)

6.8 A blogger using the pseudonym the Lame Texpatriot (2010) also took a shaded-area approach in creating diagrams for the crowds (Figures 5a and 5b). Unlike the others, however, Texpatriot chose to address the issue of crowd density in these images as well as area. Using different colors for light, moderate, and heavy densities, the map illustrates why crowd density ought to be an important consideration in these maps. The Lame Texpatriot also provided calculations on the accompanying blog post, offering multiple estimates based on different assumptions, and emphasizing that they were armchair calculations purely to argue that there could be great discrepancies in the estimates, although he drew two conclusions in his post:

There is no way the Glenn Beck rally included 300,000 people. In fact, all evidence points to it being in the range of 100,000.
Jon Stewart's rally was absolutely larger, and in fact all evidence suggests it was considerably larger, probably in the realm of 200,000+ (The Lame Texpatriot 2010).

Figure 6. Rculotta (2010a) combined photos and a map to identify areas of greatest crowd densities.

6.9 Another Internet user, going by the username Rculotta (2010a,b), attempted to capture all the strengths of the various other approaches in two displays that combined photos, perimeters, and density estimates. Rculotta's first display (Figure 6) certainly provides a unique and comprehensive entry into the collection of images in this debate. This display focuses primarily on the sizes of the densest crowds, with montages of photographs providing evidence for the map's shading choices. Moreover, Rculotta also comments that "grass showing thru indicates lower density," while still counting those areas (in favor of higher crowd size for the "Restoring Honor" rally). Though not offering a numeric estimate of attendees, Rculotta clearly tries to convey a larger, if not much larger, group at the "Restoring Sanity and/or Fear" event. Rculotta's (2010b) second display (Figure 7) incorporates the Keypro image for the Stewart/Colbert rally; it uses photos taken from different vantage points at different times to show how the crowd eventually spread into areas at the Mall's center that had been initially blocked off. A later photo (on the upper right of the display) taken from the Washington Monument makes it clear that the image used by Keypro must have been captured well before the crowd reached its maximum size.

Figure 7. Rculotta (2010b) used photos taken from different angles and at different times.

Discussion

7.1 Rculotta's displays (2010a,b) were the last – and most complex – of the images we were able to locate. What is remarkable is that they were dated November 4 – less than one week after the October 30 Stewart/Colbert rally. Compare the speed with which these claims emerged with the ten days of contentious debate in the mainstream press about the size of the Million Man March. That march was held on October 17, 1995; at the event, the organizers announced that they had drawn well over one million participants, and they responded with outrage when the National Park Service Park police later estimated the crowd at 400,000. Ten days later, researchers at Boston University offered a higher estimate (later reduced a bit after discussions with Park Service analysts) (Best 2001). In contrast, the Internet allows – and encourages – much faster responses. Individuals were posting crowd estimates and images supporting their numbers within hours after the rallies; even quite sophisticated analyses, incorporating detailed estimates of crowd shape and density, appeared within just a few days.

7.2 The Internet makes timely postings possible. The means of production – and reproduction – of claims are widely available to anyone with the necessary software savvy and access to a computer. Moreover, the speed at which these claims can be made is limited only by one's own ability to express his or her thoughts. For example, Maratea (2008) notes that the first critiques of the "Rathergate" memos were posted before Rather's program finished its broadcast. Further, there are incentives for speed. The earliest postings have increased potential to attract an audience searching for information on a topic, and search engines will often steer people toward the sites with relatively more hits. All other things being equal, an early post has a greater potential to be more influential than later ones. The fact that the Stewart/Colbert rally occurred only three days before Election Day must have added additional pressure given that national attention would shortly shift to the election results. There was an especially narrow window for capturing public attention to the crowd size debate, which may explain why Rculotta's sophisticated displays – appearing the day after Election Day – seem to have attracted little notice despite seemingly being the culmination of preceding images.

7.3 This haste may have affected the thoroughness with which the various commentators weighed evidence of comparative crowd sizes. A comparison of the images showing the areas covered by crowds (Figures 1-5b) shows considerable disagreement. For instance, Beck's crowd is variously shown as stretching to the east as far as the end of the reflecting pool (Figure 2), to the western edge of 17^th Street (Figures 3 and 5a), beyond 17^th street (Figure 4), and all the way to the Washington Monument (Figure 1). In general, commentators draw on the photo of the Beck crowd posted in Keypro's display (Figure 1)—a picture taken from the Washington Monument that showed little beyond the end of the reflecting pool. A more useful AirPhotosLive.com image, posted by CBS, appears to show a fairly dense crowd extending to 17^th street, but becoming far thinner between 17^th Street and the Washington Monument (Sundby 2010).

7.4 Similarly, the Stewart/Colbert crowd is variously shown as stretching westward roughly as far as 6^th Street (Figure 1), 8^th Street (Figure 2), 10th Street (Figure 5b), and 14^th Street (Figures 3 and 4). The most useful AirPhotosLive.com image (the bottom picture in the display in Figure 7) shows the crowd thinning out markedly west of about 11^th Street. In short, different maps depicted each crowd as having quite different dimensions. We calculate that that the most generous estimate of the east-west length of Beck's crowds was nearly 75 percent greater than the smallest estimate, and that the biggest estimate for the east-west length of the Stewart/Colbert crowd was more than 200 percent greater than the smallest estimate.

7.5 With few exceptions, the dramatic differences among the displays for the areas covered by crowds went unremarked. Maps – like statistics – tend to be treated as factual, even though the possibilities of "lying with maps" are well-known (Monmonier 1996). Although later presentations, such as those by the Lame Texpatriot and Rculotta, were more sophisticated (in terms of trying to assess crowd density and differences over time) than the earliest displays, there was relatively little commentary giving a sense that understanding of the issue was improving. If we understand science to be a cumulative enterprise, it is hard to characterize the discourse surrounding crowd sizes as popular science.

7.6 Analysts of "popular epidemiology" argue for the ability of lay movements to diagnose health problems (Brown 1992). The Internet offers a forum for all sorts of popular empirical claims – about toxic waste contamination, cancer clusters, the vaccine-autism link, and so on. The various amateur efforts to weigh in on the size of the Beck and Stewart/Colbert crowds resemble these efforts, in that they were lay attempts to make claims about empirical issues. However, perhaps because the debate lasted only a few days, a number of problems with these efforts are apparent.

7.7 First, with the exception of Rculotta, the commentators did not exploit the full array of information available. A picture taken from the top of the Washington monument has its uses, but it is especially helpful to examine other photos – especially ones that show the area closer to the Monument. Keypro's map, for instance, depicts Beck's crowd as extending far beyond the area shown in the photo included in his display. It is also clear that the time a photo was taken can make a huge difference, as evident in Rculotta's comparisons of early and later photos of the Stewart/Colbert crowd. Perhaps the shift of attention following Election Day cut the debate short, so that the discussion could not lead to a fuller analysis, but most commentators seemed ready to base their conclusions on a single image or two.

7.8 Second, there was evidence that the discussions that did occur tended to be among like-minded individuals. The thread on the conservative Free Republic site (where Figure 1 was initially posted) tended to affirm the conclusion that Beck's crowd was larger, whereas the discussion on reddit (where Figure 4 was posted) favored the view that the Stewart/Colbert crowd was larger, although minority opinions did appear on both forums.

7.9 Third, the understanding that this debate was a proxy for a larger culture war led to a variety of accusations and counter claims: Keypro's photo was not taken earlier; it was "obviously faked"; liberals invent "their own version of reality"; and so on. In other words, the Internet discussion revealed few efforts to dissect the evidence, uncover the sources of disagreement, and arrive at some shared conclusions. This is, of course, consistent with the larger history of debates about crowd sizes at demonstrations. Few are interested in crowd size as an empirical question; for most commentators, crowd size is only important as an indicator of the popularity of whatever position or interest the crowd is held to represent. If one supports Glenn Beck's positions, then claims that he drew a big crowd are welcome and must be true, and evidence that Stewart and Colbert attracted a bigger crowd is discomforting and should be dismissed – and vice versa.

7.10 Thus, the debate over the relative crowd sizes of the Beck and Stewart/Colbert rallies illustrates the potential – and the limitations – of the Internet as a forum for debating social issues. On the one hand, the Internet provided an arena where multiple, competing claims could be advanced, and in remarkably little time. Anyone who searched diligently could uncover numerous photographs and maps that offered documentary evidence, graphic interpretations, and conclusions about the relative sizes of the two rallies. On the other, the numerous claims and counterclaims were not particularly interactive; there was little in the way of dialog or even debate about the relative merits of different arguments. It is worth noting again that this case study is centered largely around a political issue. Whether the direction of discussions would be similar for other issues, such as medical or social issues, is a question that cannot be answered in the present study.

7.11 Utopian, cyber-optimistic visions of the impact of new media tend to emphasize their potential for participatory democracy, for an electronic community that can articulate differences, uncover common ground, and develop consensus around the nature of problems and their solutions. However, the Internet discussion of the crowd sizes at these rival rallies suggests that the outcomes of such debates don't look all that different from the ways claims about social problems are treated in more traditional media. Further, consistent with Maratea's (2008) analysis of claimsmaking by bloggers, the attention – or lack of attention – from those traditional media limited the impact of these claims by individuals. Even Rculotta's claims, which might have offered a means for reconciling the competing estimates, fell by the wayside.

7.12 The Internet's impact on the construction of social problems should not be exaggerated. Certainly it allows individuals to articulate claims in new ways that have the potential to reach a large, even global audience. But social problems scholars have always understood that there is a great gulf between uttering a claim and influencing social policy, a gulf that has traditionally been bridged through media coverage of claims. At least so far, the evidence suggests that new media are a valuable tool for social movements trying to mobilize support, but as a forum for claimsmaking, they continue to depend on being noticed by and chosen for coverage in established print and electronic media.

Notes

¹ Technically, the National Mall and Memorial Parks encompass more than 1,000 acres of parkland, some of it spreading out north and south of the strip between the Lincoln Memorial and the Capitol (National Park Service 2011). Some purists restrict the term "National Mall" to the strip of land between the Washington Monument and the Capitol, while denoting the area between the Lincoln Memorial and the Washington Monument as West Potomac Park. Conventional usage (which we adopt) views the Mall as the rectangle, nearly two miles long, between the steps of the Lincoln Memorial on the west, and the steps on the western side of the Capitol on the east, and between Constitution Avenue on the north, and Independence Avenue on the south. However, these north-south boundaries encompass several museums, such as the National Museum of Natural History to the north, and the National Air and Space Museum to the south. Typically, demonstrations are held in the narrower space between the fronts of these museums.

² We quote posts verbatim, regardless of unorthodox spelling, spacing, abbreviations, and punctuation.

³ It is worth noting that Keypro's photo of the Stewart/Colbert rally appears to have been taken well before the rally started. Later photographs show the crowd filling the areas in the Mall's center that had originally been blocked off and are shown as empty in the photo used by Keypro (compare with Figure 7).

⁴ This is a reference to two photos credited to AirPhotosLive.com posted by Montopoli (2010c).

⁵ Another individual reports posting a critique of the Keypro photos of the Free Republic (a conservative discussion site) on November 1, which noted :

Look at the shadows. The picture is facing directly east, and from the shadows you can clearly see that the sun was in the East, maybe about 45 degrees high in the sky. I'd guess it was taken no earlier than 9 and no later than 1030.
The Return to Sanity rally started at 12 noon, and it was indeed very full down front by 9am. (the trains were certainly at full capacity by 9am, at least the Green line was)

By the next day, the post had been removed and the individual blocked from further posting on the site (personal communication).

⁶ It is not clear why this commentator assumes that the two structures ought to appear as the same size. The Capitol building is in fact about four times wider and three times taller than the Lincoln Memorial.

Appendix: List of Image Sources

Figure 1: Aerial comparison of crowds with ovals
Author: Keypro
First known appearance: October 30, 4:24PM EDT
Original URL: <http://www.freerepublic.com/focus/f-news/2617813/replies?c=82>

Figure 2: Corrections to ovals
Author: Catholicgauze
First known appearance: October 30, 8:48PM EDT
Original URL: <http://www.geographictravels.com/2010/10/how-to-lie-with-maps-and-imagery-glenn.html>

Figure 3: Block style map
Author: Unknown
First known appearance: As early as October 31
Original URL: <http://i52.tinypic.com/10qcq2o.jpg>

Figure 4: Additional block style map
Author: Unknown
First known appearance: As early as October 31, 2:43PM EDT
Original URL: <http://i.imgur.com/AoxXh.jpg>

Figure 5: Shaded areas
Author: Lame Texpatriot
First known appearance: November 2, 9:38AM EDT
Original URL: http://lametexpatriot.blogspot.com/2010/11/sanity-rally-pt-2-attendance-size.html

Figure 6 & 7: Multiple styles
Author: Rculotta
First known appearance: November 4
Original URLs: <http://www.flickr.com/photos/rcculotta/5148863993/in/photostream/>
<http://www.flickr.com/photos/rcculotta/5149469486/in/photostream/lightbox/>

References

ANONYMOUS (2010a) Untitled image, URL (consulted Jan. 2011): <http://i52.tinypic.com/10qcq2o.jpg>.

ANONYMOUS (2010b) Untitled image, URL (consulted Jan. 2011): <http://i.imgur.com/AoxXh.jpg>.

BAKER, S. (2010) 'HuffPo: 'Rally Attendance Obliterates Turnout To Glenn Beck Rally', The Blaze (October 31), URL (consulted Nov. 2010): <http://www.theblaze.com/stories/huffpo-rally-attendance-obliterates-turnout-to-glenn-beck-rally/>.

BARBER, L. (2002) Marching on Washington: The Forging of an American Political Tradition. Berkeley: University of California Press.

BEST, J. (2001) Damned Lies and Statistics: Untangling Numbers from the Media, Politicians, and Activists. Berkeley: University of California Press.

BEST, J. 2008. Social Problems. New York: Norton.

BOLDEN, M. (2010) 'Metrorail Logs 500K+ trips Saturday', WashingtonPost.com (August 30), URL (consulted Feb. 2011): <http://voices.washingtonpost.com/dr-gridlock/2010/08/metrorail_logs_500k_trips_satu.html..

BROWN, P. (1992) 'Popular Epidemiology and Toxic Waste Contamination: Lay and Professional Ways of Knowing', Journal of Health and Social Behavior 33(3): 267-81. [doi:://dx.doi.org/10.2307/2137356]

CATHOLICGAUZE (2010) 'How to Lie with Maps and Imagery: Glenn Beck versus Jon Stewart', Geographic Travels (October 30), URL (consulted Jan. 2011) <http://www.geographictravels.com/2010/10/how-to-lie-with-maps-and-imagery-glenn.html>.

CLINTON, Hillary Rodham. (2011, February 15). Remarks on Internet Freedom. URL (consulted Nov. 2011): <http://www.scribd.com/doc/48895078/Hillary-Rodham-Clinton-Feb-15-2011>.

CNN (2010) 'Wildly conflicting reports filed about Beck rally crowd size', (August 29), URL (consulted Feb. 2011): <http://politicalticker.blogs.cnn.com/2010/08/29/wildly-conflicting-reports-about-beck-rally-crowd-size/>.

C-SPAN (2010a) 'Restoring Honor Rally', (August 28), URL (consulted Feb., 2011): <http://www.c-spanvideo.org/program/HonorRall&showFullAbstract=1>.

C-SPAN (2010b) 'Rally to Restore Sanity and/or Fear', (October 30), URL (consulted Feb. 2011): <http://www.c-spanvideo.org/program/296326-1>.

DANIELS, J. (2009). Cloaked websites: Propaganda, cyber-racism and epistemology in the digital era. New Media and Society 11: 659-83. [doi:://dx.doi.org/10.1177/1461444809105345]

DIAZ, K. (2010) 'Bachmann: 1.6 million at Beck Rally', StarTribune.com (September 1), URL (consulted Feb. 2011): <http://www.startribune.com/politics/blogs/101990463.html>.

DOUBLE DUTCH POLITICS (2010) 'Diagram Shows Rally Size Difference Between Beck / Stewart', (October 31), URL (consulted Feb. 2011): <http://www.doubledutchpolitics.com/2010/10/diagram-shows-rally-size-difference-between-beck-stewart/>.

EATON, M. 2010. "Manufacturing community in an online activist organization: The rhetoric of MoveOn.org's e-mails," Information, Communication and Society 13: 174-92. [doi:://dx.doi.org/10.1080/13691180902890125]

ELLIOTT, P. and N. Syeed (2010) 'Beck: Help Us Restore Traditional American Values', Associated Press (August 29), URL (consulted Feb. 2011): <http://abcnews.go.com/Entertainment/wireStory?id=11504195>.

FISHER, D. R., K. Stanley, D. Berman, and G. Neff. 2005. "How Do Organizations Matter? Mobilization and Support for Participants at Five Globalization Protests," Social Problems 52: 102-21. [doi:://dx.doi.org/10.1525/sp.2005.52.1.102]

GAMSON, W. and A. Modigliani (1989) 'Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach', American Journal of Sociology 95(1): 1-37. [doi:://dx.doi.org/10.1086/229213]

GITLIN, T. (1980) The Whole World Is Watching: Mass Media in the Making and Unmaking of the New Left. Berkeley: University of California Press.

HILGARTNER, S. and C. Bosk (1988) 'The Rise and Fall of Social Problems', American Journal of Sociology 94(1): 53-78. [doi:://dx.doi.org/10.1086/228951]

ITO, Mizuko et al. (2008). "Foreword." The Ecology of Games: Connecting Youth, Games, and Learning. Edited by Katie Salen. The John D. and Catherine T. MacArthur Foundation Series on Digital Media and Learning. Cambridge, MA: The MIT Press, vii–ix. doi: 10.1162/dmal.9780262693646.vii

KEYPRO (2010) Untitled display, (October 30), URL (consulted Nov. 2010): <http://www.freerepublic.com/focus/f-news/2617813/replies?c=82>.

THE LAME TEXPATRIOT (2010) 'Sanity Rally pt 2 - Attendance Size Doesn't Matter (But Mine's Bigger?)', (November 2), URL (consulted Jan. 2011): <http://lametexpatriot.blogspot.com/2010/11/sanity-rally-pt-2-attendance-size.html>.

LANGMAN, L. 2005. "From Virtual Public Spheres to Global Justice: A Critical Theory of Internetworked Social Movements," Sociological Theory 23: 42-74. [doi:://dx.doi.org/10.1111/j.0735-2751.2005.00242.x]

MARATEA, R. (2008). 'The e-Rise and Fall of Social Problems: The Blogosphere as a Public Arena', Social Problems 55(1): 139-60. [doi:://dx.doi.org/10.1525/sp.2008.55.1.139]

MCPHAIL, C. and J. McCarthy (2004) 'Who Counts and How: Estimating the Size of Protests', Contexts 3(3): 12-18. [doi:://dx.doi.org/10.1525/ctx.2004.3.3.12]

MONAHAN, B. (2010) The Shock of the News: Media Coverage and the Making of 9/11. New York: New York University Press.

MONMONIER, M. (1996) How to Lie with Maps, 2^nd ed. Chicago: University of Chicago Press.

MONTANARO, D. ( 2010) 'DomenicoNBC status at 9:07 AM Aug 28', Twitter, URL (consulted Feb. 2011): <http://twitter.com/DomenicoNBC/status/22364380399>.

MONTOPOLI, B. (2010a) 'Glenn Beck: My Rally Had "Minimum of 500,000 People"', CBSNews.com (August 30), URL (consulted Nov. 2010): <http://www.cbsnews.com/8301-503544_162-20015115-503544.html>.

MONTOPOLI, B. (2010b) 'Glenn Beck "Restoring Honor" Rally Crowd Estimate Explained', CBSNews.com (August 31), URL (consulted Nov. 2010): <http://www.cbsnews.com/8301-503544_162-20015214-503544.html>.

MONTOPOLI, B. (2010c) 'Jon Stewart Rally Attracts Estimated 215,000', CBS News.com (October 30), URL (consulted Nov. 2010): <http://www.cbsnews.com/8301-503544_162-20021284-503544.html>.

MOORE , M. (2009) 'Park Service Changes Course, Plans to Offer Crowd Estimate', USA Today (January 20): 6A.

NATIONAL PARK SERVICE (2011) National Mall and Memorial Parks, URL (consulted Feb. 2011): <http://www.nps.gov>.

RCULOTTA (2010a) 'Comparison of Stewart/Colbert & Beck Crowds', (November 4), URL (consulted Jan. 2011): <http://www.flickr.com/photos/rcculotta/5148863993/in/photostream/>.

RCULOTTA (2010b) 'Detail Far End of Stewart/Colbert Crowd', (November 4), URL (consulted Jan. 2011): <http://www.flickr.com/photos/rcculotta/5149469486/in/photostream/lightbox/>

REDDIT (2010) 'The Real Relative Sizes of the Stewart/Colbert's Rally and Glen Beck's Rally', (image). URL (consulted Feb. 2011): <http://www.reddit.com/r/politics/comments/dz38k/the_real_relative_sizes_of_the_stewartcolberts/>.

RODMAN, G. B. 2003. "The Net Effect: The Public's Fear and the Public Sphere." Pp. 11-48 in Virtual Publics: Policy and Communication in an Electronic Age, edited by B. E. Kolko. New York: Columbia University Press.

SMITH, L. and W. Melillo (1996) 'If It's Crowd Size You Want, Park Service Says Count It Out', Washington Post (October 13): A34.

SPECTOR. M, and J. I. Kitsuse. 1977. Constructing Social Problems. Menlo Park, CA: Cummings.

STANGLIN, D. and J. Durando (2010) '"Sanity" Rally Draws Tens of Thousands', USAToday.com (October 31), URL (consulted Feb. 2011): <http://www.usatoday.com/news/washington/2010-10-30-rally-stewart-colbert_N.htm>.

SUNDBY, A. (2010) 'Glenn Beck Rally Attracts Estimated 87,000', CBSNews.com (August 28), URL (consulted Feb.2011): http://www.cbsnews.com/8301-503544_162-20014993-503544.html

UNDER THE MOUNTAIN BUNKER (2010) 'Size Of The Stewart/Colbert Rally Vs. Glenn Beck Rally', (October 31), URL (consulted Feb. 2011): <http://underthemountainbunker.com/2010/10/31/size-of-the-stewartcolbert-rally-vs-glenn-beck-rally/>.

WASHINGTON METROPOLITAN AREA TRANSIT AUTHORITY (2010) 'Metro Sets New Record for Highest Saturday Metrorail Ridership', (October 31), URL (consulted Feb. 2011): <http://www.wmata.com/about_metro/news/PressReleaseDetail.cfm?ReleaseID=4717>.

WILLIAMS, B. A. and M. X. Delli Carpini. 2004. "Monica and Bill All the Time and Everywhere." American Behavioral Scientist 47:1208-30. [doi:://dx.doi.org/10.1177/0002764203262344]

ZERNIKE, K. and C. Hulse (2010) 'At Lincoln Memorial, a Call for Religious Rebirth', New York Times (August 29): A15.