The new Global Left and Transnational Social movements

Christopher Chase-Dunn, Matheu Kaneshiro, Richard Niemeyer,

Preeta Saxena, Roy Kwon and James Love

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

University of California-Riverside, Riverside, CA 92521

[v. 2-13-09 9355 words]

Land Rights Demonstrators at the World Social Forum in Nairobi, January 2007

Abstract: This paper discusses the nature of the new global Left and the network of transnational social movements based on evidence gathered from surveys conducted at two World Social Forum meetings and studies of the Internet. Surveys of attendees of World Social Forums in Porto Alegre and Nairobi have produced information about where participants live, their political attitudes and the interlinks among social movements. We also report results from studies of material published on the Internet and in the future we will use trend data on the number of Web searches to study the contours of relationships among global social movements. The network of transnational social movements is rather stable over the last five years and its multicentric structure is robust and indicates a good potential for coordinated collective action in world politics.

To be presented at the annual meeting of the International Studies Association, New York, Session on the World Social Forum, Feb. 15, 2009. This is IROWS Working Paper #48 available at https://irows.ucr.edu/papers/irows48/irows48.htm

Thanks to Christine Petit and Ellen Reese for their help on this research.

Contemporary global civil society is composed of all the individuals and groups who orient their political participation toward issues that transcend local and national boundaries and who try to link up with those outside of their own home countries in order to have an impact on local, national and global issues. The new global Left is that subgroup of global civil society that is critical of neoliberal globalization, global corporate capitalism and the exploitative and undemocratic structures of global governance (de Sousa Santos 2006). The larger global civil society also includes defenders of global capitalism and the existing institutions of global governance as well as other challengers of the current global order. The new global Left is the current incarnation of a constellation of popular forces, social movements and global political parties that have contested with the powers-that-be in the world political economy for centuries. The existing institutions of global governance have been shaped by the efforts of competing elites to increase their powers and defend their privileges, but also by the efforts of popular forces to challenge the hierarchical institutions, defend workers rights, access to the commons, the rights of women and minorities, the sovereignty of indigenous peoples and to democratize the local, national and global institutions of governance.

The evolution of the modern world-system --its processes of economic development, the rise and fall of hegemonic core powers, and the waves of globalization and deglobalization -- has been shaped by a series of world revolutions (congeries of local, national and transnational struggles and rebellions that clump together in time) since the Protestant Reformation of the 16^th century (Arrighi, Hopkins and Wallerstein 1989; Boswell and Chase-Dunn 2000). The contemporary world revolution is similar to earlier ones, but also different. The research reported in this paper is an effort to comprehend the nature of the contemporary global Left in its world historical context.

The boundaries of the progressive forces that have come together in the new global Left are fuzzy and the process of inclusion and exclusion is ongoing. The rules of inclusion and exclusion that are contained in the Charter of the World Social Forum, though still debated, have not changed much since their formulation in 2001.[1]

The new global Left has emerged as a critique of global capitalism. It is a coalition of social movements that includes reincarnations of old social movements that emerged in the 19^th century (labor, anarchism, socialism, communism, feminism, environmentalism, peace, human rights) along with movements that emerged in the world revolutions of 1968 and 1989 (queer rights, anti-corporate, fair trade, indigenous) and even more recent ones such as the slow food, global justice, anti-globalization, health-HIV and alternative media). The explicit focus on the Global South and global justice is somewhat similar to earlier incarnations of the global Left, especially the Comintern, the Bandung Conference and the anti-colonial movements. The new global Left contains remnants and reconfigured elements of earlier global Lefts. But it is a qualitatively different constellation of forces. The constituent movements have been reshaped and new technologies of communication and easier long-distance transportation have helped achieve better resolution of contradictions among the movements as well as North/South issues within movements. Growing awareness of the similarities and differences between the present period and earlier periods of world revolution have increasingly allowed the movements to better comprehend both the successes and the failures of earlier efforts and their implications for the goals, strategy and tactics of the present moment.

This paper examines the organizational space of contemporary global social movements in several ways. We are interested in the geography of participation in transnational social movements and the composition of global civil society. We are also studying the network of connections among the family of global social movements. Our study is based on two very different sources of information: surveys of attendees of two World Social Forums (Porto Alegre in 2005 and Nairobi in 2007) and studies of the number of pages published on the Internet. We discuss the methodological issues involved in these different approaches to the study of movement sizes and the structure of relations among movements as well as presenting a summary of earlier findings and our most recent results.[2]

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 the world-systems analyses of world revolutions mentioned above and Antonio Gramsci’s concepts of ideological hegemony, counter-hegemonic movements and the formation of historical blocks [Gill 2000; Carroll and Ratner 1996; and Carroll 2006a, 2006b].
From whence cometh the new global Left?

Not surprisingly, the location of the gathering has the largest effect on who attends. So when the World Social Forum was held in Porto Alegre in 2005 most of the attendees were from Brazil. More came from other Latin American countries than from other areas of the global south, but there were fewer from Africa than from South Asia, and the only attendees from the Peoples Republic of China were from Hong Kong.

Figure 1: Home places of those attending the WSF in Porto Alegre, 2005

The WSF05 was rather well attended from both the U.S. and from Europe. The sparse showing from Africa and the focus of the WSF process on the global south were the main reasons why the 2007 WSF was held in Nairobi, Kenya.

Figure 2: Home places of those attending the WSF in Nairobi, 2007

Geography and tyranny of distance (transportation costs) were also evident in the pattern of attendance at the World Social Forum in Nairobi, Kenya in 2007. The majority of attendees were from East Africa, and the attendance from the rest of Africa was better than it had been when the Social Forum meetings were held elsewhere. The U.S. and Europe continued their participation, but attendance from South Asia fell off, and from China things did not improve in terms of numbers of attendees. But both the Peoples Republic of China and the Democratic Republic of Vietnam sent delegations to Nairobi and hosted booths on the circus of Nasarani Soccer Stadium where the WSF events were held.

	Number of WSF05 Participants	Percentage of WSF05 Participants Porto Alegre	Percentage of world population in 2005	Number of WSF07 Participants	Percentage of WSF07 Participants Nairobi
Core	125	20%	13%	146	29%
Semiperiphery	451	72%	55%	78	15%
Periphery	49	8%	32%	283	56%
Total	625		6,451,392,455	557

Table 1: Residence of WSF attendees by world-system zone

             At the Porto Alegre WSF in 2005 the core was semewhat over-represented in terms of proportions of the world population (20% at the meeting but 13% of the world population,see Table 1 above). The semiperiphery was over-represented because Brazil, the site of the meeting, is a semiperipheral country and is adjacent to semiperipheral Argentina (see Appendix B for our categorization of countries into world-system zones.) The periphery, which contains 32 percent of the world’s population, was seriously under-represented at Porto Alegre (8%). As mentioned above, this was an important part of the rationale for holding the 2007 World Social Forum in Nairobi, Kenya. Table 1 also shows the distribution of attendees at the WSF07 in Nairobi across world-system zones. At the Nairobi meeting the periphery is over-represented (56%), rather than under-represented as it was in Porto Alegre, because Kenya and the surrounding countries in East Africa are in the world-system periphery. The core countries at the Nairobi meeting were even more over-represented (29%) than they had been in Porto Alegre (20%). The semiperiphery at the Nairobi meeting was seriously under-represented. Only 15% of the attendees at the WSF07 were from the semiperiphery, which has 55% of the world’s population. The 2009 World Social Forum was held in Belem, Para, Brazil.

Table 2 shows the core-semiperiphery-periphery breakdown by racial identification. These racial classifications were based on an open-ended question in which respondents were asked to identify their race or ethnicity; respondents’ answers were later recoded in terms of being “white” or “non-white.”

	WSF05 White	WSF05 Non-white	WSF07 White	WSF07 Non-White
Core	70% (49)	30% (21)	77% (106)	23% (32)
Semiperiphery	51% (158)	49% (153)	31% (22)	69 % (48)
Periphery	15% (5)	85% (29)	7% (17)	93% (246)
Total	212	203	145	326

Table 2: World-system residential zone by racial identification of attendees at the WSF05 and WSF07

Global racial stratification is reflected in the attendees at the World Social Forum meetings, as in nearly everything else. At the WSF05 seventy percent of those from the core (of those who chose to answer the question) were self-identified as white, while only 51% of those from the semiperiphery and 15% of those from the periphery were self-identified as white. At the WSF07 in Nairobi the racial breakdown was both similar and different. The percentage of whites from the core (77%) was even higher than that reported from the Porto Alegre meeting (70%). Significantly more of the Nairobi attendees from the semiperiphery were non-white compared to the WSF05 meeting (69% vs. 49%). And the proportion of whites from the periphery at the Nairobi meeting (7%) was significantly smaller than that proportion from the Porto Alegre meeting (15%). Some of these differences are due to the fact that the meeting was held in Africa. Table 2 shows that even a progressive and explicitly anti-racist effort such as the Social Forum process reflects the racial stratification of world society. Race and racism are North-South issues that must be addressed by all the counter-hegemonic movements (Starr 2004; Reese et al. 2007).

The Political Attitudes of the Global Left

Table 3 shows some of the results of our surveys at the WSF05 and the WSF07 regarding the political attitudes of attendees. While 56 percent of those at Porto Alegre wanted capitalism to be abolished (Question A), significantly fewer Nairobi respondents (only 34 percent) answered in this manner; most of the latter group sought to reform capitalism when asked to choose one answer. We also did a similar survey of attendees at the U.S. Social Forum meeting that was held in Atlanta, Georgia in the U.S. in July of 2007.[3] The Atlanta results are useful for interpreting some of the differences that we find between Porto Alegre and Nairobi.

		WSF 2005		WSF 2007		USSF 2007
A. Views on Capitalism
Reform		44%		55%		36%
Abolish		56%		34%		56%
Neither		n/a		10%		8%
B. Views on IMF
Negotiate/ Reform		14%		63%		23%
Abolish and Replace		59%		15%		21%
Abolish		27%		18%		54%
Do Nothing		n/a		4%		2%
C. Views on UN
Reform		n/a		78%		67%
Replace		n/a		10%		18%
Abolish		n/a		5%		11%
Do Nothing		n/a		7%		4%
D. Political Views
Far Left		n/a		10%		45%
Left		n/a		36%		37%
Center Left		n/a		13%		8%
Center		n/a		17%		4%
Center Right		n/a		11%		1%
Right		n/a		5%		1%
Far Right		n/a		1%		1%
Indifferent		n/a		8%		3%
E. Best Level to Solve Contemporary Problems
Community/ Sub-national provinces	59%		51%		58%
National		10%		10%		10%
International/ Global		31%		40%		32%
F. Part of Global Social Movement
No		n/a		17%		13%
Yes		n/a		83%		87%
G. Views on Establishing Democratic World Government
Good idea, and it’s possible		25%		47%		45%
Good idea, but it’s not possible		39%		38%		27%
Bad idea		36%		15%		29%
H. Views on WSF not Taking a Political Stance
Agree		46%		69%		n/a
Disagree		54%		24%		n/a
Neutral		n/a		07%		n/a
I. In Favor of Tobin Tax Proposal
No		n/a		20%		13%
Yes		n/a		80%		88%
J. In Favor of Reparations for those Affected by Slavery, Colonialism, and Racism
No		n/a		15%		07%
Yes		n/a		85%		93%
K. In Favor of Quotas to Increase Women’s Political Representation
No		n/a		14%		22%
Yes		n/a		86%		79%
L. In Favor of Women’s Right to an Abortion
No/ never		n/a		32%		13%
Yes, under all circumstances		n/a		33%		72%
Sometimes/ it depends		n/a		36%		15%

Table 3: Political attitudes of WSF attendees

The proportionally fewer supporters of abolishing capitalism at the Nairobi conference is probably due to differences in the degree and nature of government support for the meetings. In Brazil the radical goals of the World Social Forum are supported by the Brazilian Workers Party, which controlled the municipal government in Porto Alegre during the 2005 WSF meeting, and whose candidate was the president of Brazil (Ignacio Lula da Silva). The government of Kenya, on the other hand, viewed the opportunity of hosting the WSF meeting in Nairobi primarily in terms of increasing tourist visits. The Kenyan government did not support the radical and populist goals of the WSF. The whole tenor of the meeting in Nairobi was different. There were many more religious groups attending in Nairobi. The official sponsor was a cell phone company, and the admission fee was so expensive that poor Nairobi residents could not afford to attend until after they had blocked the highway adjacent to the soccer stadium where the forum was held. This difference in the nature of official support is probably the main explanation for the much less radical attitude toward capitalism at the Nairobi meeting (Reese et al 2008).

Also, a significantly greater share (63 percent) of Nairobi respondents wanted to reform rather than abolish the International Monetary Fund (IMF), compared to 14 percent of Porto Alegre respondents (Question B). And similar patterns were found for attitudes toward the World Bank and World Trade Organization (WTO) (results not shown). We think this difference is also due to the different kinds of support that the Porto Alegre and Nairobi meetings got from the local and national host governments.

Question C in Table 3 shows that the both the Nairobi and the Atlanta respondents were more favorable in their attitude toward the United Nations (UN) than in their attitudes toward the International Monetary Fund in Question B. The question about the U.N. was not asked in Porto Alegre. Only 5% of the Nairobi respondents wanted to abolish the U.N., while 18% wanted to abolish the IMF.

We did not ask Question D about Left-Right political views in Porto Alegre. There is a very large difference between Nairobi and USSF in Atlanta in the percentage of attendees who identify themselves as Far Left (10% vs. 45%). The U.S. Social Forum was not supported strongly by the government of the United States, nor by the local government, but the WSF International Council went to great lengths to find organizers who would mobilize grass roots movements within the U.S. to organize and attend the first U.S. Social Forum. This resulted in a strong showing of attendees who were both very critical of capitalism (Question A) and who identified as the Far Left. We suspect that if we had asked Question D in Porto Alegre the answers would have been more similar to the Atlanta results than to the Nairobi results. Della Porta et al.’s (2006) study of the 2002 European Social Forum provides interesting comparisons: 96 percent of respondents identified as left-of-center, with 37 percent labeling themselves as “extreme left.”

Question E asked respondents to indicate the best arena for solving the problems created by global capitalism. The alternatives presented were: community or subnational region, national, or international-global. In both surveys, when asked to choose one, more than half of the respondents chose the “community or subnational region” as the best level for addressing the problems of global capitalism (Question E). Only about 10 percent chose the national level. And the rest chose the international/global level. Question F asked whether or not respondents considered themselves to be part of a “global movement.” This question was not asked in Porto Alegre. Despite the large number of respondents who value “acting locally,” 83% of the Nairobi respondents and 87% of the Atlanta respondents claim that they are participating in the global social movement.

We asked the respondents in all three surveys their attitude toward establishing a democratic world government, and allowed three different answers (bad idea, good idea and possible, and good idea but not possible). There was significantly more support for establishing a democratic world government among the Nairobi respondents than among the Porto Alegre respondents. A large majority of respondents in both surveys believed that creating a democratic world government was a good idea (Question G); however, less than half of the Nairobi respondents and only one-quarter of WSF05 respondents, believed that this was both good and possible. Support for a democratic world government was quite high in the Nairobi sample (85 percent indicated it was a good idea). Our study of North/South differences in attitudes toward global institutions (Reese et al 2007) based on the WSF05 survey found that attendees from the Global South were much more skeptical about global institutions than respondents from the Global North. We interpreted this as probably due to the heritage of colonialism and neocolonialism. The Nairobi results seem to contradict this finding. A very large proportion of the attendees in Nairobi were from Africa, and yet the attitudes toward global institutions and their potential for democratization were quite positive. This suggests that there may be big differences by region in the peripheral s zone of the world-system. A survey of European Social Forum participants similarly found that 80 percent of respondents agreed that building new institutions of world government would be the best way to advance the “causes of the movement” (della Porta et al. 2006).

The Nairobi attendees were less supportive of the WSF taking positions on political issues (Question H) than the Porto Alegre attendees (24 percent versus 54 percent). This might be another instance in which the more conservative atmosphere in Nairobi is reflected in a less activist stance. The debate over whether or not the World Social Forum should itself become a political actor has become rather moot. Even at the Porto Alegre WSF in 2005 nearly half of the attendees supported the idea that the WSF should be an an open forum and should not itself take political positions. But the WSF Charter explicitly encourages participants to form whatever political entities they want to organize. The Assembly of Social Movements, usually held on the last day of the WSF meetings, formulates, adopts and proclaims consensual political declarations. So it is possible to both have the cake and to eat it. The World Social Forum can remain an open space, and global political organizations can be formed out of the Social Forum process.

There was a fairly high degree of consensus among respondents in our Nairobi survey on the other political questions shown in Table 3. For Question I, eighty percent of the respondents support the Tobin tax (a tax on international financial transactions that would be used to redistribute income from rich nations to poor nations). Eighty-five percent favor reparations for people adversely affected by slavery, colonialism, and racism (Question J). And for Question K, 86% favor the use of quotas by political parties and governments to increase women’s political representation. Question L, about women’s right to abortion found 35% against in Nairobi but only 13% against in Atlanta. Again we interpret this as due to differences in the nature of support for the two meetings.[4]

The Contours of the Global Network of Social Movements

At the Porto Alegre World Social Forum in 2005 and the Nairobi World Social Forum in 2007 attendees were asked with which of a long list of social movements they were actively involved. A large proportion were actively involved in more than one movement and from these choices we can see which movements share individual activists and we can infer the structure of alliances among the 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.

The extent and pattern of network linkages among the memberships and among the organizational leaderships of social movement organizations may be highly consequential. 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, gender, or issue area) may inhibit communication and make coordinated action more difficult. The ways in which social movements are linked may facilitate or obstruct efforts to organize cross-movement collective action. Network analysis can reveal whether or not the structure of alliances contains separate subsets with only weak ties, or the extent to which the network is organized around one or several central movements that mediate ties among the other movements.

The eighteen movements that we studied in 2005 are listed in Table 4 below. At the WSF in Nairobi, Kenya we used most of these same movements, but we separated human rights from anti-racism and we added nine additional movements (development, landless, immigrant, religious, housing, jobless, open source, and autonomous, and this same larger list was used at the USSF in Atlanta in July of 2007. Most of our results reported here use only the original list of 18, with human rights and anti-racism combined as they were in the 2005 survey so that we can compare the Nairobi results with the Porto Alegre results.

The WSF05 size distribution of the eighteen movements in terms of number of participants who say they are actively involved is shown in Table 4.

	number of selections	% of total selections
human rights/anti-racism	161	12%
environmental	142	11%
alternative media/culture	133	10%
peace	113	9%
socialist	87	7%
global justice	81	6%
labor	72	6%
anti-globalization	68	5%
fair trade	67	5%
feminist	66	5%
health/HIV	52	4%
indigenous	48	4%
anti-corporate	43	3%
national liberation	38	3%
slow food	38	3%
queer rights	37	3%
communist	32	2%
anarchist	20	2%
Total Responses	1298	100%
Number of Respondents	560

Table 4: Attendees Actively Involved in Social Movements, WSF05

The size distribution of the WSF05 movement selections in Table 4 shows that the highest percentages of selections were made of human rights/anti-racism (12%), environmental (11%), alternative media/culture (10%) and peace (9%). Some activists refuse to participate in the World Social Forum (or have held counter-events) and some others (e.g. those advocating armed struggle) are excluded by the WSF Charter. These factors might account for the small numbers of some of the movements (e.g. anarchists and communists). It is said that anarchists do not fill out questionnaires, but we had very few refusals and 20 of the Porto Alegre respondents and 43 of the Nairobi respondents indicated that they were actively involved anarchists.

In order to use network analysis we must choose a “cut-off” point that defines strong versus less strong ties among movement pairs. We selected a tie strength cut-off of one-half of a standard deviation above the mean number of movement interconnections to define a “strong” linkage. Using this cutoff, Figure 3 displays the “strong ties” among the movements as indicated by attendees at the WSF05.

Figure 3:The network of movement linkages at the 2005 WSF in Porto Alegre

Even with this rather high cutoff, all of the movements are connected by only one degree of separation, except for the anarchists, communists, queer rights and national liberationists. They are not connected to the network because they fall below the dichotomization point in terms of movement connections using the one-half of a standard deviation above the mean cut-off. But the matrix of all ties has no empty cells, so even these movements shared some members with other movements.

Figure 3 shows the centrality of Human Rights/Anti-Racism and Environmental movements in the network of transnational social movements represented at the World Social Forum in 2005. It also indicates that the Peace, Alternative Media, Anti-Globalization and Global Justice movements are rather central. These six movements are major“hubs” or an “inner circle”. But the overall structure is multicentric with not very large differences in network centrality among the most central six. This pattern is not very hierarchical. No single movement is the most “inclusive” or a “peak organization” for all of the others.

Figure 4: The network of movement linkages at the 2007 WSF in Nairobi

Comparison of the WSF05 and WSF07 networks (Figures 3 and 4) shows that the basic multicentric structure of the movement of movements did not change. In both networks there is a set of hub movements that strongly integrate all the other movements. The same cut-off of one-half of a standard deviation above the mean number of connections is used in both Figure 3 and Figure 4, and the same movements are used. As was the case for the WSF05 network, the matrix of movement pairs for the WSF07 meeting has no zeros, meaning that all of the 18 movements have at least one participant in all of the other movements. Only one of the forty-three Nairobi anarchists is also actively engaged in the slow food movement, the socialists and the feminists. But there are no zeros, as was also the case in the WSF05 matrix.

Three of the movements (anarchist, communist and queer rights) that were disconnected by the high bar of connectedness in the WSF05 matrix were also disconnected in WSF07, whereas National Liberation met the test in Nairobi, but not in Porto Alegre, and Anti-corporate failed the test in Nairobi but not in Porto Alegre.. One of the same movements appears near the center (Human Rights/Anti-racism), but some that were rather central in 2005 have moved out toward the edge in 2007 (Peace, Global Justice and Alternative Media). The Environmentalists are still toward the center, but not as central as they were in Porto Alegre. Health/HIV is much more central than it was in Brazil, probably reflecting both an increase in global concern and a much greater crisis in Africa. Regarding overall structural differences between the two matrices, the 2007 network is more centered around a single movement (Human Rights/Anti-racism), but there are also more direct connections among some of the movements out on the edge (e.g. feminists and socialists, socialists and labor, slow food and global justice.

The biggest finding based on the survey data is that there is a fairly stable network structure of movements as indicated by the responses of attendees at Social Forum meetings in very different locations and even though the meetings had rather different kinds of support from governments and political organizations. A somewhat similar structure is found when we examine the contents of web pages (see below). This means that there is indeed a global movement of movements and that it is a multicentric network without separate factions and integrated by a set of more central movements that link the rest. The key role played by the human rights movement in the Social Forum process is important, even though many activists resent the use of the human rights discourse by the powers that be to justify the existing structures of global governance. Those who want to democratize global governance should confront this issue head on by sharply distinguishing genuine concern for individual and group civil rights as well as issues of economic human rights and economic democracy, from the use of human rights discourse to justify neoliberal policies.

But we also found interesting differences between the network structures at the two WSF meetings studied. Human rights/anti-racism was more central at the Nairobi meeting than it was at the Porto Alegre meeting. In Nairobi human rights/antiracism was the only link between many traditional leftist movements, such as socialism, labor, and national liberation ( and also indigenism) and more middle-of-the-road or more functionally specific, and newer, social movements, whereas in Porto Alegre the older left and labor movements had direct links with the newer or more specifically functional social movements. The Kenyan government was not very sympathetic with the radical political goals of the World Social Forum. It should also be mentioned that Nairobi is a major hub for Northern NGOs operating in Africa. These contextual factors may account for the greater centrality of human rights/anti-racism at the WSF07. The alternative media group was also quite a bit less well linked to other movements at the WSF07 in Nairobi than it had been in Porto Alegre. We suspect that this may reflect the better connections that alternative media activists, who are mainly from Europe and the U.S., have with Latin America than with Africa.

So geography and the nature of local political support have important effects, and we think these are the main explanations for the differences we find between the WSF05 and WSF07 in the network of movements. We do not see any strong temporal trends in the development of the movement network based on our studies of WSF attendees.

Social Movements on the Internet

We also study the connections among movements by counting Web sites that mention pairs of movements by name. This is a different approach to studying 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.[5] 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.

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. 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 as Indicated by Web Publications

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 quotation marks return pages that have all the words together, whereas a search without the quotation marks would return all pages that contain the words even though they are at separate places in the web document. Petit then did this for each of the other sixteen movements listed in Table 5 below. And then she typed in pairs of movements e.g. “civil rights movement” “anarchist movement”.

Table 5 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 increased 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 5 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	28%	34%	31%
labor/labour	20%	16%	14%
peace/anti-war	19%	20%	12%
women's/feminist	13%	10%	11%
environmental	7%	7%	7%
socialist	3%	2%	4%
communist	2%	1%	3%
gay rights	2%	5%	2%
human rights	2%	1%	8%
anarchist	1%	1%	1%
anti-globalization	2%	1%	1%
national liberation/ sovereignty	1%	0.20%	3%
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 Number of Hits	2,055,310	39,377,900	12,556,340

Table 5: Movement sizes as indicated by relative numbers of Web page hits

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 5 are civil rights, labor, peace, feminism and environmentalism. Below these the movement page count percentages are much smaller. In Appendix A below 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.[6]

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.

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 reported above. But the distribution of counts was severely skewed, 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. 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.

                  Logged Pearson’s r correlation coefficients

2004-2006        .91

2004-2008        .91

2006-2008        .88

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, 2008

Table 5 shows the coreness scores calcutated from the dichotomized movement matrices for 2004, 2006 and 2008. Coreness reflects the density of connections of each node. Rable 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 5: Movement network structure in 2004 based on web hit pairs

Despite our assumption, mentioned in Footnote 6 above, that the human rights and civil rights movements are basically nearly the same thing, 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 6: 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 7: 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

The recent WSF meeting in Belem strongly focused on environmental issues and indigenous issues, taking advantage of its setting in the Amazon Basin. The somewhat antagonistic relationship between politicians holding office in Latin American nation-states and the activists in grassroots social movements that is enshrined in the WSF Charter reflects the autonomist and horizontalist notions that have been strong in the Social Forum process. Some observers have reported that this antagonism seems to be decreasing as the pink tide of progressive governments in Latin America has spread and deepened in some countries to include grass roots movements. Several of the populist, indigenous and Bolivarian socialist leaders affirmed many of the political themes of horizontalism and popular mobilization that have been an important part of the discourse in the Social Forum process . It is possible that a new convergence of progressive forces in the context of the rapidly developing crisis of global capitalism could result in a much more significant role for the global majority of poor and oppressed peoples in world politics. One can assume that Mao Zedong, were he still alive, would once again say that “the situation is excellent.”

References

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

__________ 2008. “Towards the fifth international?” Pp. 123-143 in Katarina Sehm-

Patomaki and Marko Ulvila (eds.) Global Political Paties. London: Zed Press.

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

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, C., 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).

Chase-Dunn, C. and Ellen Reese 2007 “The World Social Forum – A Global Party in the Making?” Pp. 53-91 in Katarina Sehm-Patamaki and Marko Ulvila (eds.) Global Political Parties. London: Zed Press.

Chase-Dunn, C. and Matheu Kaneshiro 2009 “Stability and Change in the contours of

Alliances Among movements in the social forum process” in David Fasenfest (ed.)

Engaging Social Justice. Leiden: Brill.

Chase-Dunn, C. Richard Niemeyer, Alexis Alvarez and Hiroko Inoue 2009 “Scale transitions and the evolution of global governance since the Bronze Age.” Pp. 261-284 in Wlliam R. Thompson (ed.) Systemic Transitions. New York: Palgrave MacMillan.

della Porta, Donatella. 2005. “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.

della Porta, Donatella, Massimiliano Andretta, Lorenzo Mosca, and Herbert Reiter. 2006. Globalization from Below: Transnational Activists and Protest Networks. Minneapolis, MN: University of Minnesota Press.

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

Gill, Stephen 2000 “Toward a post-modern prince? : the battle of Seattle as a moment in the new politics of globalization” Millennium 29,1: 131-140.

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

Kentor, Jeffrey 2000 Capital and Coercion: The Role of Economic and Military
Power in the World-Economy 1800-1990 . New York and London : Garland
Press.

Kentor, Jeffrey. 2005. “Changes in the Global Hierarchy, 1980-2000.” IROWS Working Paper #46

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.https://irows.ucr.edu/papers/irows25/irows25.htm

Reese, Ellen, Mark Herkenrath, Chris Chase-Dunn, Rebecca Giem, Erika Guttierrez,

Linda Kim, and Christine Petit 2007 “North-South Contradictions and

Convergences at the World Social Forum” Rafael Reuveny and William R.

Thompson (eds.) Globalization, Conflict and Inequality: The Persisting

Divergence of the Global North and South. Blackwell.

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                      “Research Note: Surveys of World Social Forum Participants Show Influence of

Place and Base in the Global Public Sphere. Mobilization 13,4 (December).

Appendix: IROWS Working Paper # 45. https://irows.ucr.edu/papers/irows45/irows45.htm

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 2008 Social Movements for Global Democracy. Baltimore: Johns Hopkins University Press.

Smith, Jackie and Dawn Weist 2006 “National and global foundations of global civil society” Pp. 289-316 in C. Chase-Dunn and S. Babones (eds.) Global Social Change. Baltimore: Johns Hopkins University Press.

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.

Starr, Amory. 2004. “How can anti-imperialism not be anti-racist?: the North American anti-globalization movement.” Journal of World-Systems Research 10 (1): 119–151, Special Issue on Global Social Movements Before and After 9-11, edited by Bruce Podobnik and Thomas Ehrlich Reifer.

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.

World Social Forum Charter http://wsf2007.org/process/wsf-charter/

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

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

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

             For Table x 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.

Appendix B: Classifications of countries into world-system zones

World Bank classification[7] World-system position[8]

Global “North:

High income

Australia Core

Austria Core

Belgium Core

Canada Core

Denmark Core

Finland Core

France Core

Germany Core

Greece Semiperiphery

Hong Kong (China) Semiperiphery

Ireland Core

Israel Semiperiphery

Italy Core

Japan Core

Korea (Rep.) Semiperiphery

Netherlands Core

Norway Core

New Zealand Semiperiphery

Portugal Semiperiphery

Spain Core

Sweden Core

Switzerland Core

Taiwan (excluded from all sources) Semiperiphery

United Kingdom Core

United States Core

Global “South”:

Upper-middle income

Argentina Semiperiphery

Chile Semiperiphery

Costa Rica Semiperiphery

Lebanon Periphery

Mexico Semiperiphery

Malaysia Semiperiphery

Panama Semiperiphery

South Africa Semiperiphery

Uruguay Semiperiphery

Venezuela Semiperiphery

Lower-middle income

Armenia Periphery

Bolivia Periphery

Brazil Semiperiphery

Colombia Semiperiphery

Dominican Republic Periphery

Ecuador Periphery

El Salvador Periphery

Iraq Periphery

Paraguay Periphery

Peru Periphery

Philippines Periphery

Low income

Bangladesh Periphery

India Semiperiphery

Kenya Periphery

Nepal Periphery

Pakistan Periphery

Sudan Periphery

Senegal Periphery

Vietnam Periphery

---

[1] The charter of the World Social Forum does not permit participation by those who attend as representatives of organizations that are engaged in, or that advocate, armed struggle. Nor are governments, confessional institutions or political parties supposed to send representatives to the WSF. See World Social Forum Charter http://wsf2007.org/process/wsf-charter/

[2] The University of California-Riverside Transnational Social Movement Research Working Group web page contains the WSF05, WSF07 and USSF survey instruments. See http://www.irows.ucr.edu/research/tsmstudy.htm

[3] We do not much discuss the Atlanta results here because we are focusing on the global Left, and think the best way to understand it is by studying gatherings that are explicitly global in intent and that try to draw participants from all over the Earth.

[4] For a more complete report on the demographic and political characteristics of survey respondents at the WSF05, the WSF07 and the USSF see Reese et al 2008 and the data appendix at https://irows.ucr.edu/papers/irows45/irows45.htm

[5] 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).

[6] The most important difference in the lists of movements between the survey and Internet page studies is the presence of “civil rights” in the Internet study and its absence in the WSF surveys. One suspects that these two involve the same people and similar issues. Table 5 above shows that civil rights is the largest single movement in our web studies, whereas human rights was a very large and central movement in the WSF survey results.

[7] Based on the Gross National Income per Capita in 2004 (World Bank 2006; see also: www.worldbank.org/data/).

[8] Based on Kentor’s measure of the overall position in the world economy in 2000 (Irows46). The cutoff point between core and semiperipheral countries has been set at 2.00, the cutoff point between semiperipheral and peripheral countries at –0.89.