Alex Wright

Anatopic Obliteration

August 7, 2006

Gartner's Nick Gall sent a few comments on my earlier posts about the obliteration phenomenon (archived here and here). He suggests that Eugene Garfield's usage of the term may be too broad, encompassing what are actually two discrete trends:

• Permanent obliteration, in which an idea's original creator is wiped out from the paper trail altogether, and
• Anatopic or palimpsestic obliteration, in which the progenitor is simply overshadowed by later, more prominent authors.

Gall thinks the latter might be better explained by the Matthew effect, a term coined by Robert K. Merton to describe the tendency of better-known scientists to receive more credit than their less well-known peers (this also strikes me as an example of power laws).

Gall writes: Merton used his term(s), not to refer to complete obliteration, but to the temporary omission of any citation and the subsequent mis-citation to the most recent author to mention a concept or use an aphorism... Thus, Merton's "Matthew Effect" would be a primary cause of "anatopic or palimpsestic syndrome" [Garfield's 2nd use], but not a cause of the permanent obliteration phenomenon [Garfield's 1st use]. That sounds about right to me. Gall also points out Stigler's Law of Eponymy, also known as the law of lesser attribution, which states that "no scientific discovery is named for its original discoverer." That law seems amply borne out on the Web as well. In my earlier post, I used the example of Ajax as an idea closely associated with a particular writer (Jesse James Garrett, to whom I certainly mean no disrespect), rather than with the engineers who actually invented the underlying technologies. Here's another example: did Tim Berners-Lee really "invent" the Web? What about Nelson, Engelbart, Bush or Otlet?

As for the two flavors of obliteration, my hunch is that on the Web, the line between these two phenomena may get impossibly blurry, insofar as the constant fluidity of Web pages makes it all but impossible to track citations with anything like the precision of Garfield's Science Citation Index (whose influence ranking method provided the direct inspiration for Google's pagerank algorithm).

These are interesting distinctions to ponder, although I don't think any of them particularly contradicts my original point: that the Web's reliance on explicit citation ranking mechanisms gives us a flawed picture of how culture really works. And as these invisible algorithms exert more and more pull over our cultural, economic and social lives, we might do well to keep their limitations in mind, and to resist putting too much faith in the purported wisdom of crowds.

File under: Informatics

_____________________
« Beyond the tag cloud | Offline »