Coding Semiperipherality: Definitional and Empirical Issues
Thomas D. Hall
Dept. of Sociology & Anthropology
DePauw University
Greencastle, IN 46135
thall@depauw.edu
web: http://fs6.depauw.edu:50080/~thall/hp1.htm
On leave through fall 06 at:
523 The Parkway
Ithaca, NY 14850
607-266-8079
Paper for presentation at Measuring and Modeling Cycles of State Formation, Decline
and Upward Sweeps since the Bronze Age, ISA Workshop presentation, March 20, 2006.
San Diego.
Working Draft
Ver: February 23, 2006
Updated March 16
NOTE: This working draft is intended to get a running start on figuring out how to
conceptualize and measure semiperipherality. It is meant as an opening bid for a
continuing conversation.
Introduction
The semiperiphery [SP] is a problematic region with problematic relations in worldsystems analysis. Even when restricted to the “modern” world-system, that is post ca
1450 C.E., it is a zone in motion and an intermediate zone which stabilizes the entire
system according to Wallerstein (1974). Several writers have tried to explicate the
complex roles of the semiperipheral regions (Arrighi 1985; Arrighi and Drangel 1986;
Chase-Dunn 1990; Hopkins et al. 1987). Even the earliest quantitative studies of worldsystem position yield problematic results, most easily resolved by considering SPs that
are rising and SPs that are falling within the world-system (Snyder & Kick 1979, see too
Smith & White 1992).
In extending world-systems analysis to precapitalist contexts Chase-Dunn and Hall
(1997, p. 78) suggest that there are at least 5 kinds of semiperipheries:
 regions that mix both core and peripheral forms of organization;
 regions spatially located between core and peripheral regions;
 regions spatially located between two or more competing core regions;
 regions in which mediating activities linking core and peripheral areas take place;
or
 regions in which institutional features are in intermediate in form between those
forms found in adjacent core and peripheral areas.
Others might be:
 regions that are rising in world-system position;
 regions that are falling in world-system postion;
1


regions that use productive technologies intermediate with respect to core and
peripheral regions;
others
Mitchell Allen (2005) and others recognize a special role for contested peripheries and by
extensions semiperipheries (# 3 above), that is areas betweeen two world-systems. La
Lone (2000) analyzes the roles of the SP in the Inka empire. Turchin (2003) and Hall
(2000; 2005a, 2005b, 2006) note that semiperipheral regions are often on or part of major
frontiers. These suggest still other characteristics of interest:
 regions that, due to various mixing and intermediate positions often have more
complex internal social structures than either core or peripheral regions;
 regions that have complex ethnic mixes;
 closely related regions that are marked by considerable ethnic dynamics,
including ethnocide and ethnogenesis;
 and combining #4 in the first list and the third in the second list, regions that
have a peculiar or special productive system that enhances their intermediate
role[s] (Central Asia nomads are the quintessential example).
Chase-Dunn and Hall (1997) argue that semiperipheral regions are often seedbeds of
system change. Briefly semiperipheral regions have sufficient information and contact
with core regions to understand assorted core processes and technologies, including
various technologies of power (Mann 1986), but remain sufficiently distant from those
technologies to consider new forms. Furthermore they have considerable interest in
developing new techniques of power and material technology to facilitate a rise in
position to core power.
In short, semiperipheral areas are very dynamic regions. Thus, even more than other
historical situations, any description or measurement of semiperipherality is in actuality a
“freeze frame” or snapshop of fluid and volatile region. This creates special measurement
issues and problems. Given this I will argue that we need more indicators of conditions in
semiperipheral regions than either core or periphery so that we may assess various
aspects of their condition[s] relevant to any analytical issue that may arise.
So the first task is to list all the characteristics embedded in the above lists. The second
task is to figure out various measures or indicators of those characteristics. The following
lists and discussions are opening round approximations that will need to be emended and
extended in light of discussions at this workshop and in subsequent work. Clearly, too,
there will need to be room for “other“ characteristics and/or detailed annotations.
Turchin’s Metaethnic Frontiers
I will not give a thorough presentation of Peter Turchin’s theory of metaethnic frontiers,
but only highlight issues that I think bear on semiperipherality – besides Peter Turchin is
major member of this group and he can correct, emend, or extend as he sees fit. I pull this
from his Historical Dymanics (2003: ch. 4).
2
Turchin seeks to explain, via empirically testable hypotheses, why frontiers between
semiperipheral and core areas are “incubators of group solidarity” (asabiya, after Ibn
Khaldun). A metaethnic frontier is one of sharp boundary difference where very different
polities and/or ethnies abutt each other. (Other important factors are frequency and
intensity of intergroup conflict and population density vs. carrying capacity). This effect
can be intensified if the metaethnic boundary coincides with a sharp geographical
boundary. He sees two problems with the discussion of semiperipherality as
conceptualized by Chase-Dunn & Hall (1997):
 too many kinds of SPs
 differentials in success of SP challengers
He argues that metaethnic frontier theory is stronger here because it sees greater potential
for successful challenge to core areas when the imperial frontier coincides with a
metaethnic frontier. We do not disagree, and find this a useful extension of our argument.
I do, however, want to add a few points that I think are germane to our task at hand. First,
as noted above there are probably many more types of SPs then the 5 we listed in 1997.
This is not a problem, but it does caution care and distinctions among them. Drawing
again on Turchin’s account, world-systems/empires have two sorts of boundaries: those
with other empires, those with nonstate groups. The same might be said of core – noncore
boundaries. What is interesting about SPs is that they have BOTH types of boundary or
frontier zones, and both are metaethnic.
The frontiers with the core or cores (depending on whether the core is uni- or multicentric) are potentially metaethnic in at least three ways. First, the core is often different
ethnically as well as politically. Furthermore, as core – SP rivalry intensifies the salience
of that frontier will increase and is very likely to become metaethnic if it was not already
so. Second, the mode of organization, and possibly technological modes [modes of
production in a non-marxian sense] are likely to be different. Third, core elites will
typically see the SP elites as inferior, as “barbarians.” (Think here of Phillip of Macedon,
Alexander’s father, calling himself “Phillip the Barbarian” viz-a-viz the Greek polises).
The frontiers with peripheral areas are of the second type, between state and nonstate
peoples. This, too, is metaethnic, first, on the basis of very different political structures.
Second, there is the “tribal zone” effect (Ferguson & Whitehead 1992a, 1992b) which
intenifies warfare among nonstate groups and between them and the SP state. Third, in
such settings raiding and trading are often alternative means – for nonstate groups – to
obtain state goods. There is a tendency for both sides to view the other as
undifferentiated. Thus states see nonstate peoples as barbarians, whereas nonstate peoples
see members of the state from peasant, to local administrator, specific city, and the entire
state as the same entity. As frontiers persist knowledge develops on each side. And, each
side uses this misunderstanding to manipulate relations. This is the stuff of any detailed
frontier history, and accounts for why even with a steep metaethnic frontier there can be
continued interactions, trade, intermarriage, alliances and so forth across the frontier (see
Brooks 2002 for detailed examples and Resendez 2005, for a discussion of how
nationality, here we could read as asabiya is constructed in such situations.)
3
For both types of metaethnic frontiers a critical, and very thorny, issue is when do these
relations build larger and stronger identities and when to they divide? This is part and
parcel of frontier ethnopolitics, which is somewhat beyond the scope of the issues at
hand. However, these issues should be held in mind as continuing problems that may
shape not only measurement issues but also analyses.
With these issues in mind I turn to characteristics of SPs.
Characteristics of Semiperipheral Regions:
Characteristics of SPs could be organized in various ways. The one proposed here is
heuristic. It could be organized in other ways. Some items repeat because they fit more
than one category.
One set of characterics revolve around the intermediate or mixed forms in many SPs:
less complex than core regions, more complex than peripheral regions:
1. regions that mix both core and peripheral forms of organization regions spatially
located between core and peripheral regions;
2. regions in which institutional features are in intermediate in form between those
forms found in adjacent core and peripheral areas;
3. regions that use productive technologies intermediate with respect to core and
peripheral regions.
Here the salient feature is social forms that less complex than those in core areas, yet
more complex than those in peripheral areas
Another set of chacteristics revolves around mediating or linking roles. This, of course,
is the classical Wallersteinian SP: C:SP::SP:P, but there are others.
4. regions in which mediating activities linking core and peripheral areas take place;
5. regions spatially located between two or more competing core regions;
6. regions that use productive technologies intermediate with respect to core and
peripheral regions;
7. regions that have a special productive system that enhances their intermediate
role[s] (Central Asian nomads are the quintessential example).
The emphasis here is on the linking either core and peripheral areas or even different
world-systems. Beaujard’s (2005) analysis of Indian Ocean trade also emphasizes these
linking roles. Those that link separate systems are ephemeral in that once such linking
takes place this defines a new, larger, merged world-system. But when the linkage is
contested the two systems often remain separate (see Allen 1997).
Both Beaujard (2005) and Allen (2005) note the roles in innovation played by SPs that
play linking roles in the larger system. Another reason SPs often are seedbeds of change,
is that in the linking role they are locus where elements developed in different parts of the
same system, or in different systems mix, facilitating various forms of hybridization and
synergy. If this is correct, we have a partial answer to Turchin’s critique: innovations are
more likely in SP regions that play linking roles in networks. The more intense this role –
that is more routes or regions linked, heavier the traffic – the more likely they are to be
seeds of invention. Indeed, one of Beaujard’s examples (2004, p. 443) is the invention of
4
an alphabet in Ugarit & Palestine 13th c BCE. With more areas interacting with different
languages, the more important record keeping becomes, and record keeping systems that
can be used in more than one language – that is, they are sound based as in syllabaries of
alphabets – the more efficient they will be. This suggests likelihood, but not fully explain
why in some SPs and not in others.
In the third and fourth entries in this category (three repeats from the first category) the
emphasis is on the linking role, where the productive technology serves as a means of
connecting core and peripheral regions, or separate core regions. The difference between
these two is that in the first the productive technology is “between” the types found in
core and peripheral areas, such as mixed swidden and hunting that link agricultural
production in core areas with foraging in peripheral areas. In the second example, Central
Asian nomads, the productive technology is not intermediate in any sense, pastoralism is
quite distinct (Barfield 1993), and allows the linking of multiple core areas.
Still another set of characteristics are dynamics, in particular whether the specific SP is
rising or falling within the world-system.
8. regions that are rising in world-system position;
9. regions that are falling in world-system postion;
10. regions that are marked by considerable ethnic dynamics, including ethnocide and
ethnogenesis.
These are examples where SP region is more volatile than either core or peripheral areas.
Finally, there are all the rest, the residual categories:
11. regions that, due to various mixing and intermediate positions often have more
complex internal social structures than either core or peripheral regions;
12. regions that have complex ethnic mixes.
This category repeats some of the earlier entries, but here the emphasis is on complex
social structures that are the result of combining BOTH core and peripheral elements, but
not necessarily as an intermediate form, but as a complex combination.
Given the overlapping of categories and multiple roles of some characteristics how might
these be assessed?
Measures and/or Indicators of Semiperipherality
Based on the preceding discussion[s], I will propose – again as an opening gambit – some
possible measures [M] or indicators [I] of SP categories and characters.
The first set is composed of basic indicators of semiperipherality:
1. regions that mix both core and peripheral forms of organization regions spatially
located between core and peripheral regions;
2. regions in which institutional features are in intermediate in form between those
forms found in adjacent core and peripheral areas;
3. regions that use productive technologies intermediate with respect to core and
peripheral regions.
5
Here two indicators or measures are needed. The basic ones would be some sort of
“measure” of complexity in terms of social organization, e.g. class; institutions such as
succession rules for leaders, form of elite identification, and so on; and productive
technology, e.g. agriculture with fields over some preset size; presence of merchant
guilds, etc.
Second would be indicators of relative complexity with respect to core and peripheral
regions.
The second set is composed of indicators of mediating or linking roles:
4. regions in which mediating activities linking core and peripheral areas take place;
5. regions spatially located between two or more competing core regions;
6. regions that use productive technologies intermediate with respect to core and
peripheral regions;
7. regions that have a special productive system that enhances their intermediate
role[s].
For 4, presence of entrepôt-like centers, cities or ports that have traffic BOTH to core and
pereripheral areas, or between separate core regions. Notation should indicate whether
that is the only path or one of several. For instance, for much of the first millennium CE
the silk road states are all links between separate core regions, and there is no by-pass
around them that is direct China to Rome. Some of the South Asian maritime states might
receive the same coding.
In a recent article Beaujard (2005) analyzes Indian Ocean trade. He notes that “In the
fifteenth century—after 1433 to be precise--when China, along with its huge market,
practically cut itself off from foreign trade, commerce was thereby disrupted in the China
Sea, but in Southeast Asia commerce continued to flourish, and Malacca was described
by Portuguese at the start of sixteenth century as ‘the biggest wharehouse in the world.”’
On the following page (436) he observes that links to China & India, … ”more easily
compensate for recessions…” That is from sometime late in the first millennium CE the
oceanic circuits were in some sense alternate routes to the overland Silk Roads. Beaujard
argues, ***”But the flourishing of the Silk Roads and the route across the Mongol
steppes in the thirteenth century did not weaken trade in the Indian Ocean” (2005, p.
447).
All this suggests that we need to note the number of circuits in which a SP is involved,
and the number of different core regions with which it trades.
5. This could be a simple binary code: 1 if between two core areas; 0 if between core and
periphery or core and hinterland.
6. This would be a relative complexity measure as in 3. We need to develop some more
elaborate conceptualiztion of just what we mean by intermediate technology, something
6
like processing ore into ingots, or repackaging ingots, but not making final goods, such as
swords or plows.
7. Again we need do some conceptual work here, but this could for starters be simple
binary: 1 is distinctive technology that is neither core nor peripheral (e.g. nomads); 0 if it
is a version of core or peripheral technologies.
For dynamics obviously we need some sort of time frame. This would need to be
specified in advance for consistency.
8. regions that are rising in world-system position;
9. regions that are falling in world-system postion;
10. regions that are marked by considerable ethnic dynamics, including ethnocide and
ethnogenesis.
We might use a version that Wilkinson has used several times. A three-way
categorization: +1 for rising/increasing complexity; 0 for relatively stable, and -1 for
falling or declining. For world-system postion rising/falling/stable are the categories. For
ethnic composition they would be increasing complexity, decreasing complexity, stable.
We might want separate indicators for ethnogenesis and/or ethno- or geno-cide.
*IF* data is sufficiently refined we might even scale degree of rising and falling, say
rapidly and slowly as categories. Again we would need to specify cut points.
Finally, there are the residual categories:
11. regions that, due to various mixing and intermediate positions often have more
complex internal social structures than either core or peripheral regions;
12. regions that have complex ethnic mixes.
For 11 we might have three binary variables:
A. 0 if social structure is not unique
1 if social structure is specific to that SP
B. 0 if core elements are not present
1 if core elements are present
C. 0 if peripheral elements are not present
1 if peripheral elements are present
For 12 there could be two different measures:
One would be the number of ethnic groups that are recognized. There is a potential
problem when states lump all “tribals” as barbarians. For instance for Nuevo Mexico in
New Spain’s administration in the 17th & 18th centuries, Pueblo groups were each
individual recognized, some non-sedentary peoples were recognized, but often they were
lumped under categories like indios bárbaros, or norteños (barbarian indians, peoples of
the north; see Weber 2005 for many detailed examples). However, this might be useful
for the second measure, change. Not only would this include increasing number of
groups, but increasing differentiation of groups, so that, again from Nuevo Mexico, los
indios bárbaros are differentiated into Apache, Apache de Navajo, Utes, Comanches, and
so on. A similar instance would be Romans differentiating among various Germanic
groups, and no longer just lumping them under Germanii. The point here being to capture
7
increasing complexity along frontier zones, allowing us to gain indicators of metaethnic
frontiers.
Closing Comments
Obviously, the preceding discussion is far from final. There are more unresovled
problems than answers, but I think approaching it this way we can capture both the
fluidity and complexity of SP regions. This is necessary if we are to be able to explain
why some SPs are seedbeds of change, and others are not.
REFERENCES
Allen, Mitchell. 1997. Contested Peripheries: Philistia in the Neo-Assyrian WorldSystem. Unpublished Ph.D. dissertation, Interdepartmental archaeology program, UCLA.
_____. 2005. “Power is in the Details: Administrative Technology and the Growth of
Ancient Near Eastern Cores.” Pp. 75-91 in The Historical Evolution of World-Systems,
edited by Christopher Chase-Dunn and E. N. Anderson. New York and London:
Palgrave.
Arrighi, Giovanni. 1985. Semiperipheral Development: The Politics of Southern
Europe in the Twentieth Century. Beverly Hills: Sage.
Arrighi, Giovanni and Jessica Drangel. 1986. "Stratification of the World-Economy: An
Explanation of the Semiperipheral Zone." Review 10:1(Win.):9-74.
Barfield, Thomas J. 1993. The Nomadic Alternative. Englewood Cliffs: Prentice-Hall.
Beaujard, Philippe. 2005. “The Indian ocean in Eurasian and African World-Systems
before the Sixteenth Century.” Journal of World History 16:4(Dec.):411-465.
Brooks, James F. 2002. Captives and Cousins: Slavery, Kinship, and Community in the
Southwest Borderlands. Chapel Hill: University of North Carolina Press.
Chase-Dunn, Christopher. 1988. "Comparing World Systems: Toward a Theory of
Semiperipheral Development." Comparative Civilizations Review 19(Fall):29-66.
_____. 1990b. "Resistance to Imperialism: Semiperipheral Actors." Review
13:1(Winter):1-31.
Chase-Dunn, Christopher and Thomas D. Hall. 1997. Rise and Demise: Comparing
World-Systems. Boulder: Westview Press.
Chase-Dunn, Christopher, Daniel Pasciuti, Alexis Alvarez and Thomas D. Hall. in press.
Growth/Decline Phases and Semiperipheral Development in the Ancient Mesopotamian
and Egyptian World-Systems. Pp. 114-138 in Globalization and Global History edited
by Barry K. Gills and William R. Thompson. London: Routledge.
8
Ferguson, R. Brian, and Neil L. Whitehead, eds. 1992. War in the Tribal Zone:
Expanding States and Indigenous Warfare. Santa Fe, NM: School of American
Research Press.
______. 1992. "The Violent Edge of Empire. Pp. 1-30 in R. B. Ferguson and N. L.
Whitehead, eds. War in the Tribal Zone. Santa Fe, New Mexico: School of American
Research Press.
Hall, Thomas D. 2000. "Frontiers, Ethnogenesis, and World-Systems: Rethinking the
Theories." Pp. 237-270 in A World-Systems Reader: New Perspectives on Gender,
Urbanism, Cultures, Indigenous Peoples, and Ecology edited by Thomas D. Hall.
Lanham, MD: Rowman & Littlefield.
_____. 2005a. Central Asia, Nationalism, and Globalization: Past Lessons for Future
Possibilities, paper presented at Central Eurasian Studies Society meeting, Boston, Sept.
29-Oct. 2, 2005.
_____. 2005b. “Frontiers as Zones of Interaction: Finding the Buddha in the Shard Pile,”
paper presented at American Society for Ethnohistory meeting, Santa Fe, November 19.
_____. in press. Review of Changing National Identities at the Frontier: Texas and
New Mexico, 1800-1850 by Andrés Reséndez. Forthcoming in Pacific Historical Review.
_____. 2006. “[Re] Periphalization, [Re]Incorporation, Frontiers, and Nonstate Societies:
Continuities and Discontinuities in Globalization Processes.” Pp. 96-113 in Globalization
and Global History edited by Barry K. Gills and William R. Thompson. London:
Routledge.
Hopkins, Terence K., and Wallerstein, Immanuel. 1987. "Capitalism and the
Incorporation of New Zones Into the World-economy." Review 10:4(Fall):763-779.
La Lone, Darrell. 2000. "Rise, Fall, and Semiperipheral Development in the Andean
World-System." Journal of World-Systems Research 6:1(Spring):68-99 [ejournal
http://jwsr.ucr.edu/index.php].
Mann, Michael 1986. The Sources of Social Power: A History of Power from the
Beginning to A.D. 1760. Cambridge: Cambridge University Press.
Reséndez, Andrés. 2005. Changing National Identities at the Frontier: Texas and New
Mexico, 1800-1850. Cambridge, Cambridge University Press.
Smith, David A. and Douglas White. 1992. “Structure and Dynamics of the Global
Economy: Network Analysis of International Trade 1965-1980.” Social Forces
70:4(June):857-893.
Snyder, David and Edward L. Kick. 1979. "Structural Position in the World System and
Economic Growth, 1955-1970: A Multiple-Network Analysis of Transnational
Interactions." American Journal of Sociology 84 (March):1096-1126.
Turchin, Peter. 2003. Historical Dynamics. Princeton, NJ: Princeton University Press.
9
Wallerstein, Immanuel. 1974. The Modern World-System: Capitalist Agriculture and
the Origins of European World-Economy in the Sixteenth Century. New York:
Academic Press.
Weber, David J. 2005. Bárbaros : Spaniards and Their Savages in the Age of
Enlightenment. New Haven: Yale University Press.
10