Social movement Networks

As Reflected in Web Publications

Christopher Chase-Dunn and James Love

Department of Sociology and Institute for Research on World-Systems (IROWS)

University of California-Riverside, Riverside, CA 92521

 Draft v. 4-2-09 6133 words

 

Abstract: This paper uses the Internet to study the sizes of and the contours of relationships among contemporary social movements. Trends in movement size are estimated and we use published web pages that contain mentions of pairs of social movements to examine the network of movements. These results are compared with other studies that use survey research to study movement networks. We use both counts of published web pages produced by Google searches and trends in the volume of searches produced by Google Trends.

To be presented at the annual meeting of the Pacific Sociological Association, This is IROWS Working Paper #49 available at http://irows.ucr.edu/papers/irows49/irows49.htm

Thanks to Christine Petit and Richard Niemeyer for their help on this research.

 

            This paper examines the organizational space of contemporary social movements by counting Web sites that mention pairs of movements by name.  In this study we seek to understand the structure of connections among progressive social movements and how those connections may be evolving over time. For this purpose we analyze results obtained from using the Google search engine to count the number of web sites containing certain phrases and pairs of phrases.[1] We examine the contours of the social movement connections found on web pages, and how these have changed over time.  The tricky problem is that our findings probably reflect other things as well as changes in the structure of the network of popular movements in the global public sphere. Undoubtedly our choice of the English language and the vagaries of Internet search engines may also be consequential for our findings. We shall try to sort out these different elements affecting the structure of movement connections based on Web publications.  

            There is a large scholarly literature on networks, coalitions and alliances among social movements (e.g. Carroll and Ratner 1996; Krinsky and Reese 2006; Obach 2004; Reese, Petit, and Meyer 2008; Rose 2000; Van Dyke 2003). Our study is theoretically motivated by this literature as well as by world-systems analyses of world revolutions (Arrighi, Hopkins and Wallerstein 1989; Boswell and Chase-Dunn 2000) and Antonio Gramsci’s analysis of ideological hegemony, counter-hegemonic movements and the formation of historical blocks [see also Carroll and Ratner(1996) and Carroll (2006a, 2006b)].

            The Internet is both a vast trove of information and a global interaction network for producing new information about how people think and their capabilities for transnational collective action. For those social scientists who are interested in global phenomena this makes feasible the extension of older research methods into new realms, and this extension is fraught with issues that need to be resolved so that we can know how much to trust the information generated by Internet research (e.g. see Bainbridge 2009). This paper will explore some of these issues.

            We assume that the number of published sites on the Web is to some extent a positive function of the number of people who support each movement. This should be the case because popular support should provide more resources and more activity, resulting in more publication. But there are undoubtedly other factors besides popular support. Wealthy individuals or political groups can pay for the production of Web publications. Official government agencies and political campaigns can use the discourse of social movements. Nevertheless we still think it is likely that the amount of publication on the Web should be related to the size and strength of social movements and we have used the Google search engine since 2004 to determine the number of web pages on which certain phrases that indicate content relative to each of seventeen social movements.

            Social movement organizations may be integrated both informally and formally.    Informally, they are connected by the voluntary choices of individual persons to be active participants in multiple movements.  Such linkages enable learning and influence to pass among movement organizations, even when there may be limited official interaction or leadership coordination.  In the descriptive analyses below, we assess the extent and pattern of informal linkage by ascertaining the links of movements as indicated by Web publications.  At the formal level movement organizations may provide legitimacy and support for one another, and they may collaborate in joint action.  The extent of identification and formal cooperation among movements both causes and reflects informal connections among participants and the publication of web documents that combine movement discourses.

            The extent and pattern of linkages among the memberships and among the organizational leaderships of social movement organizations are highly consequential for the potential for collective action in local and global political struggles.  Some forms of connection [e.g. “small world” networks, (Watts 2003)] allow the rapid spread of information and influence; other forms of connection (e.g. division into “factions” by region, culture, language or issue area) may inhibit communication and make coordinated action more difficult. The ways in which social movements are linked (or not linked) may facilitate or obstruct efforts to organize cross-movement collective action. Network analyses can reveal whether or not the structure of alliances contains separate subsets with only weak ties, and the extent to which the network of movements is organized around one or several central movement nodes that mediate ties among the other movements.

            Our Internet research on the structure of alliances among social movements has been paralleled by a series of studies of transnational social movements that have participated in the World Social Forum process (Smith et al 2007; Chase-Dunn et al 2007). These studies obtained survey responses from attendees at World Social Forum meetings in Porto Alegre, Brazil, Nairobi, Kenya and Atlanta, Georgia asking each respondent to indicate those movements in which they were actively involved.  Despite problems of comparability due to differences in the movement lists we will compare our Web-based results with those obtained from the WSF surveys.

 

The Relative Sizes of Social Movements

          Christine Petit (2004) described her Google searchs performed on July 28, 2004 as follows:  “…I typed in the phrase ‘civil rights movement’ and noted the number of websites containing that text.” The parentheses return pages that have all the words together, whereas a search without the parentheses would return all pages that contain the words even though the are at separate places in the web document. She then did this for each of the other sixteen movements listed in Table 1 below. And then she typed in pairs of movements e.g. “civil rights movement” “anarchist movement”.

             

            Table 1 shows the total number of hits for all the movements and the percentages for each movement for 2004, 2006 and 2008. It should be noted that the total number of hits increase from 2,055,310 in 2004 to 39,377,900 in July of 2006, and then decreased to 12,556,340 in October of 2008. We suspect that the Google search engine methodology became more selective between 2006 and 2008. It is quite unlikely that the number of pages on the Internet decreased. But Table 1 shows that the relative percentages of many of the movements are quite stable between 2006 and 2008, so we think it is likely that whatever difference in search engine methodology that was implemented between 2006 and 2008 did not much affect the relative distribution of movement presences on the Web, which is the focus of our study.

 

Movement

July 28, 2004

July 18, 2006

October 22, 2008

civil rights

27.80%

34.30%

30.82%

labor/labour

19.50%

15.80%

14.38%

peace/anti-war

18.60%

20.20%

12.21%

women's/feminist

12.90%

10.00%

11.29%

environmental

7.10%

7.20%

7.10%

socialist

2.50%

2.40%

4.14%

communist

1.90%

1.10%

3.20%

gay rights

1.80%

4.60%

1.82%

human rights

1.80%

0.90%

7.67%

anarchist

1.20%

1.00%

1.36%

anti-globalization

1.50%

0.70%

0.83%

national liberation/

sovereignty

1.10%

0.20%

2.92%

fair trade/trade justice

0.70%

0.40%

0.46%

global justice

0.60%

0.30%

0.35%

slow food

0.50%

0.50%

1.11%

indigenous

0.40%

0.30%

0.33%

anti-corporate

0.10%

0.00%

0.02%

Total %

100%

100%

100%

Total Number of Hits

2,055,310

39,377,900

12,556,340

                Table 1: Movement sizes as indicated by relative numbers of Web pages

            These movement categories were originally designed by Christine Petit for her 2004 study, and in order to be able to study changes over time we have held to her original categories. Civil rights in the United States is associated with the struggle for racial equality in the 1950s and 1960s. It has rather different connotations in other countries. Nevertheless the “civil rights” movement has by far the largest presence on the Web and this presence is rather stable. Indeed it increased from 28% to 34% between 2004 and 2006 and then dropped back to 31% in 2008.

            The big five movements in Table 1 are civil rights, labor, peace, feminism and environmentalism. Below these, the movements are much smaller. In the Appendix we have a table that shows movement sizes based on our studies of attendees at the World Social Forum in Porto Alegre in 2005. These results are only partly comparable because of differences in the movements that were included, but there are nevertheless important similarities in the two sets of results and this supports the validity of using web page counts to estimate movement sizes.

            We have already mentioned changes in the total number of hits between 2004 and 2008 and the up-down trend of the civil rights movement. Seven of the seventeen movements are quite stable over time in terms of relative percentages: feminism, environmentalism, anarchism, fair trade, global justice, and the indigenous movement. Peace and gay rights demonstrated an up-and-down pattern across the three time points, as did civil rights. Labor went down, as did anti-globalization, fair trade, global justice and anti-corporate. Socialism and communism went up in the last period, as did human rights and slow food.

            Can we simply equate the relative sizes as an indicator of movement strength, and can we interpret increases and decreases as changes in the popularity and power of movements? It is well-known that Internet usage and the publication of Web materials has been increasing geometrically since the mid-1990s, and that both usage and publication have been diffusing across the globe (Zook 2005). The growth and the trajectory of diffusion probably affect the distribution of movement hits differently. Some movements were early adopters and some have come later to Web publication. Thus trends should reflect the diffusion process as well as changes in movement strength. Late adopters should increase their percentages of Web publications over the time period studied and early adopters should go down in terms of the relative percentages of web pages.

            But seven of the movements did not much change their relative scores, and three others went up and then down. Labor, anti-globalization, fair trade, global justice and anti-corporate went down. Were these early adopters or did they actually decrease in movement strength relative to other movements? Socialism, communism, human rights and slow food went up in the last period. Were these late adopters or did they increase in relative movement strength?

            The geographical pattern of diffusion may also have affected the Web activity of movements. As mentioned above, the civil rights movement has been connected with the campaign for racial equality in the U.S. And the U.S. was the region of the world where Web usage and publication got its start and increased early. Per capita Internet usage is well-correlated with GNP per capita in cross-national comparison, but there is another factor as well. Those places that are far from the centers of power and information are often early-adopters of inexpensive techniques of long-distance communication. Many small towns in rural regions of the Global South now have their Internet café. One way that activists from the Global North have been able to help movements from the Global South is in establishing new web sites. But there is still a “digital divide” in Internet access, and this continues to be part of the causation of the relative distribution of movement publications.

 

Movement Topic Size as Reflected By Web Search Activity

            We have also used Google Trends, a tool for estimating the number of searches performed by the Google Search Engine, and comparing these over time to look for trends. We used this tool to examine the size relationship among movement topics for the largest movements in our study above. The number of searches is a different kind of indicator from the number of web pages published. It shows how much interest the public has in topics and how this has changed over time. Google Trends uses weekly data on searches since January of 2004.

            When we submit whole phrases such as “anarchist movement” in Google Trends  the program states that: "Your terms - "anarchist movement" - do not have enough search volume to show graphs." Only “civil rights movement” has enough search volume to return a graph. This shows that searches for civil rights movement varied over time and that the relative volume of searches tended to decline over the period from 2004 to 2009 (See Figure 1).

Google Trends scaling

According to Google, data are standardized based on the average search traffic of the term you enter.  Rather than producing raw search results for the term, Google averages the weekly search counts during the selected period and denotes this average as a 1.  For example, if we searched “civil rights” from 2004 to 2009 the graph would produce a baseline average for all searches requested during the given period, represented as 1.  However, if we observed that in early 2005, the graph spiked to 3.5, this would inform us that searches during early 2005 were 3.5 times greater than the average searches attempted from 2004-2009.  Google calls this relative scaling. When multiple search targets are included in the same graph the user is allowed to specify which of the items will be used to scale the rest of the items. We chose to standardize the search counts on the “civil rights” movement topic. The Google methodology does not make it clear how searches are compared across different languages.

 

           

Figure 1: Relative volume of searches for "civil rights movement", 2004-2009[2]

           

For our study of the relative size of movements we want to compare search volumes for the different movements. We were not able to do this for the whole phrases containing the word “movement,” but we were able to do it for main topics of the five largest movements listed in Table 1 above based the sizes of published web material. That is, we submitted the topics “civil rights,” “labor,” “peace,” “feminism,” and “environmentalism” to Google Trends to examine how the search volumes for these movement terms compare with one another. This worked in Google Trends and produced the results shown in Figure 2.

Figure 2: Relative volume of  searches for movement topics, 2004-2009

Figure 2 uses the civil rights as unity (1) to scale the relative volumes of the five movement topics. Civil rights is the blue line with diamonds and produces a trend similar to the one shown in Figure 1 except that now we can see the relative sizes of the search volumes for the other movement topics.

            Whereas the study of numbers of web pages for social movements discussed in Table 1 reveals that the civil rights movement is considerably larger than the labor movement based on the number of web pages published, when we examine the volume of searches we find that the topic of “labor” is far larger than is the topic of “civil rights” based on comparing the number of web searches by individuals using the Google search engine. So in relative size terms numbers of web publications and the volume of searches are somewhat different animals. The question for this paper is what do these indicators have to do with the relatives sizes of social movements in terms of numbers of adherents and other resources?  We surmise that the web page counts are better indicators of relative movement size than are the search volumes. A large number of web searches that contain a particular movement topic may only mean that something has happened that has piqued the public’s interest in that topic.  It is more akin to popularity than movement size. Of course popularity can be a resource. But it may also be a liability. It all depends on the meaning context. Searches for labor or civil rights may be launched by both proponents and opponents of the the associated movements.

            Nevertheless it is interesting to compare the search volumes displayed in Figure 2. By far the highest are “labor,” which is cyclical with annual high peaks in August or September, and annual low points in December or January. The “peace” topic is similar in volume to labor, but without the annual spiky peaks and low points. Civil rights is much lower in volume, but is still higher than feminism, which is itself higher than environmentalism.  It is probably unfair to use the topic of “environmentalism” to assess the amount of public interest in environmental issues.  When we substitute “global warming” for “environmentalism” in Figure 3 we see that global warming starts off at about the same level as civil rights, but then in late 2006 it shows a large wave of search volume and stays fairly high until 2008. In 2009 the wave has fallen off somewhat, but it is still twice as high as civil rights.
 

Figure 3: Movement topic search volumes with global warming, 2004-2009Figure 3: Movement topic search volumes with global warming, 2004-2009

We also use evidence from the Internet to study the interconnections among different social movements.

Network Structure of Social Movements in Internet Pages

            We compare the overlapping of movements by counting web pages that mention the names of movement pairs, e.g. environmental movement/labor movement at three time points: 2004, 2006 and 2008. The counts of overlaps need to be dichotomized for the purposes of network analysis using UCINet and we tried to do this in a way that would make our results comparable with the study of movement networks based on survey results. But the distributions of raw counts were severely skewed (e.g. see Table 2 shows the raw counts for 2008), thus raising the average of the counts far above the median of counts. To make the distribution more normally distributed we performed a logarhythmic transformation (log to the base ten) on the raw counts. We dichotomized the logged overlap counts using the same cutting point used in the network analysis reported above: ½ standard deviation above the mean to produce a network matrix of zeros and ones. Table 3 shows the dichotomized scores for 2008.

Movement

total

civil

labor

anti-war

women's

enviro.

socialist

commun.

gay

human

anti-glob

anarchist

national

global

slow food

indigen.

fair

trade just.

sovereign

anti-corp

civil rights

4,190,000

 

111,000

37,100

53,500

37,200

8,200

5,790

21,900

13,300

2,820

1,930

1,290

1,940

474

686

279

76

719

133

labor

1,150,000

111,000

 

13,400

18,800

20,700

89,300

38,800

2,620

7,120

3,170

6,270

2,470

1,890

184

800

576

63

214

211

anti-war

655,000

37,100

13,400

 

6,590

7,100

8,610

3,850

3,690

2,690

4,340

2,260

1,680

3,580

98

349

408

63

222

157

women's

787,000

53,500

18,800

6,590

 

11,400

4,240

4,840

5,980

13,800

1,250

1,030

777

685

145

559

150

22

123

53

environmental

861,000

37,200

20,700

7,100

11,400

 

2,330

1,660

1,920

15,300

3,750

1,160

208

1,170

851

679

645

70

280

229

socialist

533,000

8,200

89,300

8,610

4,240

2,330

 

12,500

2,040

1,140

2,240

5,990

1,620

530

2,530

324

30

991

856

46

communist

408,000

5,790

38,800

3,850

4,840

1,660

12,500

 

433

527

861

2,410

6,230

315

86

233

15

7

83

27

gay rights

228,000

21,900

2,620

3,690

5,980

1,920

2,040

433

 

3,260

272

149

30

67

22

19

18

5

50

7

human rights

943,000

13,300

7,120

2,690

13,800

15,300

1,140

527

3,260

 

790

90

754

596

60

425

99

10

2,350

6

anti-globalization

106,000

2,820

3,170

4,340

1,250

3,750

2,240

861

272

790

 

2,000

1,210

3,190

401

240

719

150

140

627

anarchist

167,000

1,930

6,270

2,260

1,030

1,160

5,990

2,410

149

90

2,000

 

525

471

8

73

17

7

88

28

national liberation

314,000

1,290

2,470

1,680

777

208

1,620

6,230

30

754

1,210

525

 

95

6

111

9

5

68

6

global justice

45,400

1,940

1,890

3,580

685

1,170

530

315

67

596

3,190

471

95

 

28

168

288

85

71

96

slow food

138,000

474

184

98

145

851

2,530

86

22

60

401

8

6

28

 

7

212

13

124

5

indigenous

42,100

686

800

349

559

679

324

233

19

425

240

73

111

168

7

 

34

7

108

5

fair trade

44,000

279

576

408

150

645

30

15

18

99

719

17

9

288

212

34

 

814

8

20

trade justice

15,100

76

63

63

22

70

991

7

5

10

150

7

5

85

13

7

814

 

2

6

sovereignty

52,100

719

214

222

123

280

856

83

50

2,350

140

88

68

71

124

108

8

2

 

1

anti-corporate

5,570

133

211

157

53

229

46

27

7

6

627

28

6

96

5

5

20

6

1

 

 

Table 2: Raw number of paired hits for social movements in 2008

 


 

movement

civil

labor

anti-war

women's

enviro.

socialist

commun.

gay

human

anti-glob

anarchist

national

global

slow food

indigen.

fair

trade just.

sovereign

anti-corp

civil rights

 

1

1

1

1

0

0

1

0

0

0

0

0

0

0

0

0

0

0

labor

1

 

0

1

1

1

1

0

0

0

0

0

0

0

0

0

0

0

0

anti-war

1

0

 

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

women's

1

1

0

 

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

environmental

1

1

0

0

 

0

0

0

1

0

0

0

0

0

0

0

0

0

0

socialist

0

1

0

0

0

 

0

0

0

0

0

0

0

0

0

0

0

0

0

communist

0

1

0

0

0

0

 

0

0

0

0

0

0

0

0

0

0

0

0

gay rights

1

0

0

0

0

0

0

 

0

0

0

0

0

0

0

0

0

0

0

human rights

0

0

0

0

0

0

0

0

 

0

0

0

0

0

0

0

0

0

0

anti-globalization

0

0

0

0

0

0

0

0

0

 

0

0

0

0

0

0

0

0

0

anarchist

0

0

0

0

0

0

0

0

0

0

 

0

0

0

0

0

0

0

0

national liberation

0

0

0

0

0

0

0

0

0

0

0

 

0

0

0

0

0

0

0

global justice

0

0

0

0

0

0

0

0

0

0

0

0

 

0

0

0

0

0

0

slow food

0

0

0

0

0

0

0

0

0

0

0

0

0

 

0

0

0

0

0

indigenous

0

0

0

0

0

0

0

0

0

0

0

0

0

0

 

0

0

0

0

fair trade

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

 

0

0

0

trade justice

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

 

0

0

sovereignty

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

 

0

anti-corporate

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

 

Table 3: Paired hits dichotomized at 1 standard deviation above the mean in 2008

The QAP routine in UCINet was used to calculate the Pearson’s r correlation coefficients for the logged network matrices scores in 2004, 2006 and 2008.

2004-2006        .91

2004-2008        .91

2006-2008        .88

Table 4: QAP Pearson's r correlations for logged pair scoresLogged Pearson’s r correlation coefficients

The three matrices of log scores are highly correlated with one another, indicating a rather high level of stability of the network of social movements.

                                                                                                                                            2004 2006    2008  

                   

  1             civil rights movement   .367  .336    .363   

  2             labor movement/labour   .367  .408    .393

  3           peace/anti-war movement   .367  .479    .366

  4         women's movement/feminist   .324  .297    .291

  5            environmental movement   .274  .274    .308

  6                socialist movement   .299  .310    .363

  7                communist movement   .236  .206    .291

  8               gay rights movement   .171  .146    .310

  9             human rights movement   .199  .146    .155

 10       anti-globalization movement   .270  .224    .166

 11                anarchist movement   .278  .224    .053

 12      national liberation movement   .173  .116    .000

 13           global justice movement   .143  .178    .000

 14                slow food movement   .028  .000    .022

 15      anti-corporate movement        .024  .000    .096

 16               indigenous movement   .000  .000    .000

 17 fair trade/trade justice movement   .000  .000    .000

 18              sovereignty movement   .000  .043    .155

 

Table 5: Multiplicative coreness scores from binary movement matrices, 2004, 2006 and 2008

Table 5 shows the coreness scores calculated from the dichotomized movement matrices for 20-04, 2006 and 2008. Coreness reflects the density of connections of each node. This table shows that most of the movements do not change their centrality or lack of it much over the four year time period. The gay rights movement gets more central between 2006 and 2008 and the anti-globalization movements declines since 2004. The anarchist movement become less central after 2006 as do the global justice and national liberation movements. The sovereignty movement gets more central.

            Figure 5 below depicts the network structure of movement links based on the Internet page counts for 2004.  None of the movements were completely disconnected from the matrix by the dichotomization of the logged overlap scores. But the slow food movement has only one tie to the network through the anti-globalization movement. The civil rights movement is both the largest in terms of individual counts of web pages and in terms of overlaps with other movements as shown in Figure 5.

Figure 4:  Movement network structure in 2004 based on web hit pairs

            Figure 4 includes both the civil rights movement and the human rights movement. It is often assumed that the human rights and civil rights movements are basically the same thing, but there are important differences between the two based on our web page study. The civil rights movement is far larger in terms of total web page publications than is the human rights movement. And with regard to position in the network of social movements, the civil rights movement is much more central. In retrospect it was probably a mistake to not to include civil rights in our list of studied movements at the World Social Forum. There we found that human rights/anti-racism was one of the largest movements and was the most central in the network of movements. We do not know how our study of movements using survey results at the WSF meetings might have been different if we had included civil rights.

Figure 5: Movement network structure in 2006 based on Internet page pair mentions

The 2006 web page network results included a huge jump in the total of web pages with movement name pairs, which raised the mean number of overlaps. Despite that we logged the counts, the cutting point at ½ a standard deviation above the mean left five movements disconnected from the main network of movements: slow food, indigenous, fair trade/trade justice, and anti-corporate. The gay rights movement is less well-connected than it was in 2004.

Figure 6: Movement network structure in 2008 based on Internet page pair mentions

By 2008 the slow food movement has become reconnected to the main network and the gay rights movement has again become as strongly connected with other movements as it was in 2004.

Conclusions

          As with our other studies of movement network structures based on survey research responses, the networks show a rather consistent pattern of both movement size distributions and the network of connections among movements.  In all cases we have a single multicentric web in which a few more centrally located movements connect most of the rest to one another. This is a robust network structure that is unlikely to experience major splits despite that often large contradictions among the goals pursued by the individual movements. The web network results confirm our findings based on individual commitments to movements in that there are at least some links among most of the movements. This structure bodes well for the emergence of a new global left that may be able to effectively contend in world politics.

           


Appendix: Comparing Web and Survey Results

            Christine Petit (2004) conducted a Google search engine project to study networks among social movements as represented by texts available on the World Wide Web in 2004 and in 2006 she replicated her study in order to make it possible to ascertain change over time and so that we can compare the results with our survey evidence from the Porto Alegre World Social Forum of 2005.

             

A

B

C

D

E

F

 

July 04

Web hits

% of total

July04

%movement

selections at WSF05

July06 web hits

% of total July06

% 04-06 change in hits

anarchist movement

25,100

1.7%

1.50%

395,000

1.5%

0%

anti-corporate movement

1,780

.1%

3%

15,100

.05%

-.05%

anti-globalization movement

30,300

2%

5%

291,000

1%

-1%

global justice movement

11,500

.8%

6%

112,000

.4%

-.4%

human rights movement

36,500

2.5%

12%

362,000

1.4%

-1.1%

communist movement

40,000

2.7%

2%

425,000

1.6%

-1.1%

environmental movement

146,000

10%

11%

2,820,000

11%

1%

fair trade/trade justice movement

14,830

 

1%

5%

159,200

.6%

-.4%

gay rights movement

37,100

2.5%

3%

1,830,000

7%

4.5%

indigenous movement

8,090

.5%

4%

120,000

.5%

0%

labor movement/labour

400,000

27%

6%

6,220,000

24%

-3%

national liberation/sovereignty movement

21,610

 

 

1.5%

3%

87,600

.3%

-1.2%

peace/anti-war movement

382,000

26%

9%

7,950,000

31%

5%

slow food movement

10,500

.7%

3%

199,000

.8%

.1%

socialist movement

52,000

3.5%

7%

952,000

4%

.5%

women's movement/feminist

266,000

18%

5%

3,940,000

15%

-3%

total

1,483,310

 

 

25,877,900

 

 

Table 6: Internet hits in 2004 and 2006 compared with movement sizes obtained from survey questionnaires at the World Social Forum in 2005

            For Table 6 we combined the fair trade movement and trade justice movement web hits to make the Petit study comparable with the WSF survey, and did the same with national liberation movement and the sovereignty movement. 

             The comparison between web hits and movement choices at the WSF (Columns B,C and E) show that the relative sizes are rather similar for ten of the sixteen movements that are compared. This establishes a baseline of comparability between these two very different sources of information about movement linkages.

            Six of the movements display what appear to be significant differences between web texts and numbers of activists at the World Social Forum. Human rights, global justice, indigenous and fair trade are better represented at the WSF than on the web.  Labor, peace and feminism are significantly less represented at the WSF than on the web. 

            Looking at the change scores for the web hits in Column F, we see that the biggest increases are in gay rights (4.5%) and the peace movement (5%). The women’s movement and the labor movement have gone down by 3%, but the rest of the movements have stayed about the same in percentage terms while the total numbers of hits increased dramatically between 2004 and 2006. The general stability of the relative sizes despite the rapid growth over the two year period and fairly good match with the WSF survey data increases our confidence that we are measuring something significant about the discursive space of transnational movements with the Internet results.

References

Amin, Samir 1997 Capitalism in an Age of Globalization. London: Zed Books.

__________ 2006. “Towards the fifth international?” Pp. 121-144 in Katarina Sehm-Patomaki and Marko Ulvila (eds.) Democratic Politics Globally. Network Institute for Global Democratization (NIGD Working Paper 1/2006), Tampere, Finland.

Arrighi, Giovanni, Terence K. Hopkins, and Immanuel Wallerstein.  1989.  Antisystemic MovementsLondon: Verso.

Bainbridge, William S. 2009 “Expanding the use of the Internet in religious research” Review of Religious Research

Borgatti, S. P.  and Everett, M. G.  and Freeman, L. C. 2002 UCINET 6 For Windows: Software for Social Network Analysis  http://www.analytictech.com/

Boswell, Terry and Christopher Chase-Dunn. 2000. The Spiral of Capitalism and Socialism: Toward Global Democracy. Boulder, CO: Lynne Reinner.

Byrd, Scott C. 2005.  “The Porto Alegre Consensus: Theorizing the Forum Movement.” Globalizations. 2(1): 151-163.

Carroll, William K. and R.S. Ratner 1996 “Master framing and cross-movement networking in contemporary social movements” Sociological Quarterly 37,4: 601-625.
Carroll, William K. 2006a “Hegemony and counter-hegemony in a global field of action” Presented at a joint RC02-RC07 session on alternative visions of world society, World Congress of Sociology, Durban, South Africa, July 28.
______________ 2006b “Hegemony, counter-hegemony, anti-hegemony” Keynote address to the annual meeting of the Society for Socialist Studies, York University, Toronto, June. Forthcoming in Socialist Studies, Fall, 2006.
Chase-Dunn, Christopher, Christine Petit, Richard Niemeyer, Robert A. Hanneman and Ellen Reese 2007 “The contours of solidarity and division among global movements” International Journal of Peace Studies 12,2: 1-15 (Autumn/Winter)

della Porta, Donatella.  2005a. “Multiple Belongings, Tolerant Identities, and the Construction of ‘Another Politics’: Between the European Social Forum and the Local Social Fora,” Pp. 175-202 in Transnational Protest and Global Activism, edited by Donatella della Porta and Sidney Tarrow.  Lanham: Rowman & Littlefield Publishers, Inc.

Fisher, William F. and Thomas Ponniah (eds.).  2003.  Another World is Possible: Popular Alternatives to Globalization at the World Social ForumLondon: Zed Books.

Hanneman, Robert and Mark Riddle 2005.  Introduction to Social Network Methods.  Riverside, CA:  http://faculty.ucr.edu/~hanneman/

IBASE (Brazilian Institute of Social and Economic Analyses) 2005. An X-Ray of Participation in the 2005 Forum: Elements for a Debate Rio de Janeiro: IBASE http://www.ibase.org.br/userimages/relatorio_fsm2005_INGLES2.pdf

Krinsky, John and Ellen Reese. 2006. “Forging and Sustaining Labor-Community Coalitions: The Workfare Justice Movement in Three Cities.”  Sociological Forum 21(4): 623-658.

Obach, Brian K. 2004 Labor and the Environmental Movement: The Quest for Common Ground. Cambridge, MA: MIT Press.

Petit, Christine. 2004. “Social Movement Networks in Internet Discourse” Presented at the            annual meetings of the American Sociological Association, San Francisco. IROWS         Working Paper #25.http://irows.ucr.edu/papers/irows25/irows25.htm

Reese, Ellen, Christine Petit, and David S. Meyer.  2008 “Sudden Mobilization: Movement Crossovers, Threats, and the Surprising Rise of the U.S. Anti-War Movement.”  Social Movement Coalitions, edited by Nella Van Dyke and Holly McCammon

Reese, Ellen, Christopher Chase-Dunn, Kadambari Anantram, Gary Coyne, Matheu

            Kaneshiro, Ashley N. Koda, Roy Kwon, and Preeta Saxena 2008                              “Place and Base: the Public Sphere in the Social Forum Process” IROWS Working            Paper # 45.

Reitan, Ruth 2007 Global Activism. London: Routledge.

 Rose, Fred 2000 Coalitions Across the Class Divide: Lessons from the Labor, Peace, and Environmental Movements. Ithaca, NY: Cornell University Press.

Santos, Boaventura de Sousa 2006 The Rise of the Global Left. London: Zed Press.

Sen, Jai and Madhuresh Kumar with Patrick Bond and Peter Waterman 2007 A Political Programme for the World Social Forum?: Democracy, Substance and Debate in the Bamako Appeal and the Global Justice Movements. Indian Institute for Critical Action : Centre in Movement (CACIM), New Delhi, India & the University of KwaZulu-Natal Centre for Civil Society (CCS), Durban, South Africa. http://www.cacim.net/book/home.html 

Schönleitner, Günter 2003 "World Social Forum: Making Another World Possible?" Pp. 127-149 in John Clark (ed.): Globalizing Civic Engagement: Civil Society and Transnational Action. London: Earthscan.

Smith, Jackie, Marina Karides, et al. 2007 The World Social Forum and the Challenges of Global Democracy. Boulder, CO: Paradigm Publishers.

Starr, Amory.  2000. Naming the Enemy: Anti-corporate Movements Confront Globalization. London: Zed Books.

Van Dyke, Nella 2003  “Crossing movement boundaries: Factors that facilitate coalition protest by American college students, 1930-1990.” Social Problems 50(2):226–50.

Watts, Duncan 2003.  Six Degrees:  The Science of a Connected Age.  New York:  W.W. Norton.

Zook, Matthew A. 2005 The Geography of the Internet Industry. Malden, MA: Blackwell.

 



[1] This builds on earlier research that used the number of web sites to study the sizes of and connections among social movements. See Petit (2004) and Chase-Dunn et al (2007).

[2] The graph labeled “news reference volume” in Figure 1 reflects the number of times the chosen search term appeared in Google News stories.  Google News, like other web-based search engine news sources, such as Yahoo!, reflect a synthesis, gleaning articles from a myriad of sources.  These include articles from the associate press, New York Times, and others, which are published throughout the globe.  In this sense, the News Reference Volume graph reflects the prevalence of the search term in global, media publications.  When Google Trends detects a rise in news story volume, based upon the selected search term, it creates the graph and displays Google News stories written near the time of that spike that contain the selected search term – the selected articles only contain the term in the headline of the story and not simply in the text.  Currently, only English-language headlines are displayed, but Google is currently working to expand the scope of news stories selected to include non-English media sources.