Categorization and the Nature of Science

Does the Core and the Gap exist? That is, does a generally well-off realm known as the Functioning Core contain goods associated with globalization (wealth, peace, etc), while a realm known as the Non-Integrating Gap lack these goods?

Mountainrunner, surprisingly, appears to say the answer is no. While he does not say so directly, he notes that (in general) anything that exists in the Gap exists within the Core, and vice versa. In response to my claim that the assassination of former Pakistani Prime Minister Bhutto is not surprising because it happened in the Gap, Mountainrunner wrote:

My point is this: this is about violence and death and ideology that is not specific to Islamists or the Gap. Heads of state were targeted. The IRA reminded Thatcher they only needed to be lucky once, she needed to be lucky all the time. Italy, Greece, Hungry, etc. Take your pick and you’ll find attacks on leadership.

A similar mistake is made by those who deny the existence of “race,” or often the “generations of warfare:” a misunderstanding about what categories are.

Categories are not Platonic ideals, true forms that are immutable through time and space. Categories allow us to explain variance. That is, a categorization is useful if categories within it correlate with clusters in frequencies of some traits. Consider again the Core and the Gap. There are rich people in the Gap, and poor people in the Core. There are those with IPTV in the Gap, and those without electricity in the core. For that matter, there are gang rapes in Darfur and gang rapes in Dallas.

But the terms “Core” and “Gap” really do explain variation in these things. Don’t believe it? Run the numbers yourself. The same is true with regard to race, and my assumption is that the same is true with regards to the generation of war.

Objecting to the Core/Gap categorization because you can cite assassinations in Core countries is like objecting to the concept of race because you know some East Asians who drink milk. (Because I am unaware of any large dataset on war that’s been used to test xGW theory, sadly we’re still in the realm of mixed methods when it comes to the generations of war.)

The moral of the story: categories explain variation. That does not mean they explain all variation. That does not mean they are supposed to.

Anything else is just… unscientific.

Also on the web: Against xGW, for William Lind? On another aspect of Mountainrunner’s post.

3 thoughts on “Categorization and the Nature of Science”

  1. MR,

    “The kinetic nature of the first attack on her life was indicative of the kinetics more likely to be found in the Gap, but neither her death nor the tactics in the successful attack are.”

    Is “Assassinations of major political figures are not more likely to happen in the Gap than the Core” the point you are making?

    “I didn’t see a rebuttal from you on the severity of attacks in a core country like Mexico that would break your scientific analysis.”

    How would you measure this? By percentage of recent leaders who died from political violence? Percentage removed by a coup?

    “You should be more precise in your analysis while at the same time understanding that buckets, or bins, of categorization are more than pourous, but blended in with others when you look at the details.”

    Agreed, of course. Such is trivially true. The traits the categories describe are continuous (if not smoothly continuous).

    MR + DHM,

    “Categorization is also linked I presume to our favorite topic of late: working memory. If we couldn’t think in terms of buckets (bins if you prefer), we’d be overwhelmed with complexity.”

    “The devil is certainly in the details and makes for messy analysis but have the ability to enlighten us to different trajectories, and thus counters and solutions. Overly broad statements for convenience mask the facts, limiting choices, and confuse unnecessarily.”

    Both of these statements are true, and they complement each other.

  2. Nice post, Sir. Categorization is also linked I presume to our favorite topic of late: working memory. If we couldn’t think in terms of buckets (bins if you prefer), we’d be overwhelmed with complexity.

  3. Dan, you completely misconstrue our discussion. I do not say the core-gap model doesn’t exist. Our discussion at my blog is about your statement that Bhutto’s assisination in a Muslim country shouldn’t be a surprise, a statement you base on the Core-Gap model. What comes of Bhutto’s death will be indicative of it happening in the Gap. The kinetic nature of the first attack on her life was indicative of the kinetics more likely to be found in the Gap, but neither her death nor the tactics in the successful attack are. This is the “core” of the argument.

    I didn’t see a rebuttal from you on the severity of attacks in a core country like Mexico that would break your scientific analysis.

    You should be more precise in your analysis while at the same time understanding that buckets, or bins, of categorization are more than pourous, but blended in with others when you look at the details.

    The devil is certainly in the details and makes for messy analysis but have the ability to enlighten us to different trajectories, and thus counters and solutions. Overly broad statements for convenience mask the facts, limiting choices, and confuse unnecessarily.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>