Operationalizing the Gap

Force Structure Will Change,” by Thomas Barnett and Henry Gaffney, Proceedings, October 2000, pp 30-34, http://www.thomaspmbarnett.com/published/forcestruc.htm

A Hammer Looking for Nails: The Gap, the Core, and the Final Frontier,” interview with Thomas Barnett, Raeson, 1 November 2004, http://www.thomaspmbarnett.com/interviews/RaesonInterview.pdf.

Viral in-coring: Seoul to Beijing,” by Thomas Barnett, Thomas P.M. Barnett :: Weblog, 4 January 2006, http://www.thomaspmbarnett.com/weblog/archives2/002774.html.

http://www.thomaspmbarnett.com/glossary.htm,” Thomas P.M. Barnett, downloaded 8 April 2006, http://www.thomaspmbarnett.com/glossary.htm.

In this post I will try to put together an operationalization and some alternate rival hypotheses for Tom Barnett’s PNM Theory.


I need to finish a research design for my Scopes & Methods class. The rough draft was on traditional geopolitics, but needed considerably work. I kicked around ways to to save it, yet I had trouble focusing on writing that just doesn’t matter. I learn so much more from blog writing than class writing that I find myself looking forward to typing in new posts, but assignments are drudgery.

Until the obvious hit me: write it as a blog post! It’s not a good blog post — it’s actually the perfect combination that doesn’t work either as a tdaxp post or as something I could hand in — but at least it gets me motivated. So today’s work discusses the Research Question, Independent Variable, Dependent Variables, and Alternate Research Hypotheses required to operationalize the Gap.

I would also like to acknowledge the work of Catholicgauze, Chicago Boys, Coming Anarchy [1, 2, 3] and The Glittering Eye in “mapping the gap.” Those posts were inspirational.

That’s easy enough to say. What’s hard is writing something in a somewhat similar style to everything else here. But I’ll try.

Research Question

“Thomas Barnett’s description of “life in the Gap” accurately predicts whether a country is in the Core or the Gap. That is, as the degree to which a state matches the criteria increases, it is more likely to be categorized as “Gap.” Further, the same is true if you look at a state’s neighbors with the same criteria. Further, the same is true if one divided the world into “Old Core,” “New Core,” and “Gap.” Further, the same is true if you look at a state’s neighbors with that criteria. Further, all of these relationships are more valid than the description’s correlation with G77 membership, “first,” “second” and “third” world membership, or categorization by the United Nations Human Development Index. If this is the case, Barnett’s “new map” is preferable to the popular models of the Global North and the Global South.”

Independent Variable

The independent variable (IV) is comprised of measures of life in the Gap’s poverty, nastiness, shortness, brutality., and solitariness, as described on pages 161-166 of Barnett’s The Pentagon’s New Map

The Gap Quotient

  • “Life in the Gap is poor”
    “Of the 118 countries listed by the world Bank as ‘low-income’ or ‘low-middle income’ (below $2963 per capita annual), 109 are located in the Gap.”
  • “Life in the Gap is nasty”
    “According to Freedom House’s 2003 survey of states around the world, 48 out of a global total of 192 surveyed were rated as ‘not free.’ Of those 48, 45 are located within the Gap.
  • “Life in the Gap is short”
  • “Of the 50 states with the highest life expectancy rates (76 to 83 years), four-fifths lie within the Core. However, if we are to look at 50 states with the lowest expectancy rates (37 to 57 years), all but one (South Africa) lie within the Gap.

  • “Life in the Gap is brutal”
    “No matter what list of “current conflicts” you want to work from (e.g. University of Maryland … ) you’ll come up with a number somewhere short of three doze, with 80 to 90 perfect of them falling squarely inside the Gap.”
  • “Life in the Gap is solitary”
    “A good measure of communications connectivity today is the number of Internet hosts found in a country. No surprise here: the more developed your economy becomes, the more connected your people become.”

Calculation for each portion of the value will be in the form “(x – min(x)) / (max(s) – min(x))” which will produce a 0 to 1 value for each state. After this, the five components (poverty, etc) will be averaged for a final 0 to 1 value. A higher value indicates more poverty, nastiness, shortness, brutality, and solitariness.

Dependent Variables

The Dependent Variables (DVs) will be a numeric score of how well the state actually falls into Barnett’s categories. Both a simple Core-Gap and a more complex model will be used.

DV1: Simple

Two possible values, Core or Gap, will be used. Core will have a value of “1,” while Gap will have a value of “2” Determining where a state falls is as simple as looking at PNM‘s inside back cover and seeing what side of the “Boundary of the Non-Integrating Gap” a state falls in. Alternatively, one can get the map off the web.

DV1 is expected to rise as the IV rises.

DV2: Simple Neighbors

Barnett focuses on geographic proximity in his definition of the Gap

Today, the Non-Integrating Gap is made up of the Caribbean Rim, Andean South America, virtually all of Africa, portions of the Balkans, the Caucasus, Central Asia, the Middle East, and most of Southeast Asia. These regions constitute globalization’s “ozone hole,” where connectivity remains thin or absent in far too many cases. Of course, each region contains some countries that are very Core-like in their attributes (just as there are Gap-like pockets throughout the Core defined primarily by poverty), but these are like mansions in an otherwise seedy neighborhood, and as such are trapped by these larger Gap-defining circumstances.

So a test should be run to see if this is a factor. The DV2 for a state will be the average of the DV1s for each of its neighbors.

DV2 is expected to rise as the IV rises.

DV3: Complicated

The more complicated version of this will differentiate the Old Core from the New Core. The Old Core, which is “anchored by America, Europe, and Japan” but “excludes South Korea,” by Old Core would seem to be

  • The United States of America
  • Canada
  • Belgium
  • France
  • Italy
  • Luxembourg
  • Netherlands
  • Germany
  • Denmark
  • Ireland
  • UnitedKingdom
  • Greece
  • Portugal
  • Spain
  • Austria
  • Finland
  • Sweden
  • Norway
  • Switzerland
  • Iceland
  • other Euro micro states

with all other “Core” states as New Core. In this version, “Old Core” is 1, “New Core” is 2, and “Gap” is 3.

DV3 is expected to rise as the IV rises.

DV 4: Complicated Neighbors

DV4 will be calculated relative to the DV3s in the same way that DV2 are calculated with DV1, for the same reason.

DV4 is expected to rise as the IV rises.

DV 5: Not Pursued

Even less defined than “Old Core” and “New Core” is Barnett’s concept of “Seam States.” His glossary defines them as:

The countries that ring the Gap–such as Mexico, Brazil, South Africa, Morocco, Algeria, Greece, Turkey, Pakistan, Thailand, Malaysia, the Philippines, and Indonesia. Some are already members of the Core, and most others are serious candidates for joining the Core. These states are important with regard to international security, because they provide terrorists geographic access to the Core. The U.S. security strategy regarding these states is simple: get them to increase their security practices as much as possible and to close whatever loopholes exist.

I don’t think that Barnett offers a coherent definition here, so I will let this one pass.

Alternative Rival Hypothesis

A research design should have one of more alternate rival hypothesis. After all, perhaps IV and DV increase together –but an even better correrlation could be found by using something else.

It struck me that I could keep my generally geographic theme by seeking to compare Tom Barnett’s Core and Gap model with the earlier Global South model.

The Global North-Global South certainly is a good alternate rival hypothesis, because it’s one that Barnett sees as similar to but different from his. To Raeson Nyhedsmagasinet:

I studiously avoid the “North-South” concept since I have my ABC’s in South America [Argentina, Brazil, Chile], like Australia and South Africa are in the South. I try to focus on: who’s connecting up?

Yet it’s one he and Hank Gaffney previously used (in describing possible force structure strategies):

This camp sees the main foreign policy task of the next decade being the processing of Russia and China into the great power fold on our terms meaning they learn to play by our rules. Once the North is in order, the South should fall in line, especially since the rogues would not have anyone of consequence to supply them in their nefarious activities.

… [or] …

This camp sees the main foreign policy task of the next decade being the effective management of the economic and technological gaps dividing North and South. You keep the North s economic expansion on track by making sure nothing and no one in the South messes it up. When situations down there get really ugly, you do what you have to, but you avoid serious involvement unless key economic fault lines are involved.

… [or] …

The South needs help now, and if it does not get it, it will bring its pain to us one way or another. Slowing down globalization s march also will give much-needed breathing space to the New Economy s losers in the North (e.g., low-tech labor).

Perhaps Barnett isn’t describing a Core-Gap conflict at all, but just a North-South conflict. If this is the case, the specific lines Barnett draws are more a geostrategic convenience for American interests than lines that match what is happening on the ground.

ARH 1: G77


Perhaps instead of Gap, the best fit for Barnett’s Hobbesian places are the members of the G77. In this ARH, G77 would be “1” and states not in the G77 would be two.”

ARH 2: The First, Second, and Third World


In the Three Worlds model, the First World would be “1,” the Second World would be “2” and the Third World would be “3”

ARH 3: UN Human Development Index


The UNHDI is the most rigorous of the atlernate rival hypothesis, because it is also based on statistical information. Because like the IV in this experiment is it comprised of objective data of health, it should be strongly correlated with the IV.

This paper proposes that the Research Hypothesis of the validity of Barnett’s “new map” is true, and that all dependent variables proposed are better than any of the alterate research hypotheses. If this is not true — then perhaps we better stick to the “old map”!