The Age of Intellectual Property?

Are we in the Age of Intellectual Property?

It’s become a truism in IP scholarship to introduce a discussion by acknowledging the remarkable recent rise in popular, scholarly, and political interest in our field. Thus readers will recognize a familiar sentiment in the opening line of Amy Kapczynski and Gaëlle Krikorian’s new book:

A decade or two ago, the words “intellectual property” were rarely heard in polite company, much less in street demonstrations or on college campuses. Today, this once technical concept has become a conceptual battlefield.

Only recently, however, has it become possible to put this anecdotal consensus to empirical test.

In December 2010, Google launched ngrams, a simple tool for searching its vast repository of digitized books and charting the frequency of specific terms over time. (It controls for the fact that there are many more books being published today.)

If you haven’t already played around with this tool to explore your own topics of interest, you should. While you’re at it, take a stab at explaining why writing on the Supreme Court rose steadily until approximately 1935 and has dropped just as steadily ever since!

Back to our topic, though. What does this data reveal about the prominence of intellectual property in published discourse?

I generated two graphs, both charting the terms “intellectual property,” “copyright,” “patent,” and “trademark.” First, the longview:

Relative frequency of IP terms in American English since 1800

Since 1800, patents have been a far more popular topic of conversation and study than trademark and copyright law. Perhaps not surprisingly, given the persistent popular interest in science and invention.

References to patents peaked quite dramatically in 1890. (The exact date can be pinpointed if you remove smoothing from the graph, visit here to manipulate it yourself.) This pique/peak of interest coincides with the founding of General Electric (based on Thomas Edison’s lightbulb patents), the passage of the Sherman Antitrust Act, and the 100th anniversary of the first US patent statute.

For much of these two centuries, reference to “intellectual property” – the conceptual grouping of patents, trademarks, and copyrights taken so much for granted today – was virtually nonexistent. That really changes only in the past few decades. Here’s the view since 1960:

Shaver ngram of IP terms since 1960

Relative frequency of IP terms in American English since 1960

A few changes stand out over these five decades. Patents finally lose their position of prominence as the American public takes increasing interest in copyright. Not surprisingly, this coincides with the emergence of the Internet.

Ayn Rand

The more dramatic shift, however, is the change in fortunes of the term “intellectual property.” What we’re seeing here is the invention of a new concept. People had long written about patents, copyrights, and trademarks. But the idea of “intellectual property” was a new one.

One of its earliest appearances is a 1967 Ayn Rand piece (Alan Greenspan is listed as an editor). Rand sought to dispel the traditional conception of patents and copyrights as privileges granted by the government, positioning them instead as quintessential examples of property.

The remarkable success of this idea – in culture and in the law – is self-evident today.

Compared to the last 200 years, the last fifty reveal the dramatically increasing importance of copyright and trademark in public conversation. (If the public is perhaps less interested in patents today than a century ago, interest in securing them certainly continues to rise.)

Even more important than the rising importance of these individual legal terms, a new concept has emerged that underlies their moral justification, empirical study, and legal regulation. That change is what defines the Age of IP.

This idea not only opens the conversation in Kapczynski and Krikorian’s book, but underlies the very concept of the project, as underscored by the title.

The editors and contributors refuse the historically naive view that would take the concept of “intellectual property” as a given. Instead, the contributions situate IP historically, as a contested concept that various social movements seek to destabilize.

One last graph. This one uses the rising use of “intellectual property” since 1960 as a baseline to plot increasing interest in the terminology of the A2K movement: words like “public domain,” “open access,” “fair use,” and of course, “access to knowledge.”

Relative frequency of A2K terms since 1960

One way to interpret this graph is as suggesting that the term “access to knowledge” (the royal blue line) is today where the term “intellectual property” (the sky blue line) was in 1975.

Unfortunately, Google’s ngram only makes available data up to 2008. Kapczynski and Krikorian’s Access to Knowledge in the Age of Intellectual Property doesn’t yet show up on this chart. You can download the full text for free, however, from Zone Books and MIT Press.

(Also missing from the chart is Access to Knowledge in Egypt, published in 2010, thanks to Bloomsbury Academic and my incredible coeditor Nagla Rizk. “Mabrouk!” Nagla, on rejoining the Internet today. I’ll keep following your updates on the Egyptian revolution via Twitter.)

You may also like...

3 Responses

  1. Maryland Conservatarian says:

    Wow – an Ayn Rand reference without snark or a sneer. Nice.

    Yesterday (2/2/11) would have been her 106th birthday.

  2. Frank Pasquale says:

    Very interesting points. Post-Rand, I think there is also a counter-movement in libertarian thought to consider IP as a form of government intervention. See, e.g.,

    “Several writers, Stephan Kinsella most notably among them, have argued that patents and copyrights should not form part of a proper libertarian law code. These writers modify and extend the work of Murray Rothbard, who allowed copyrights but not patents.”

    Dean Baker pushes this viewpoint for the libertarian left.

    I personally have found Tom W. Bell’s conception of copyright as author’s welfare ( and the creep toward “regulatory copyright” rather appealing. I believe that, the more one looks at the details in the copyright code, the more it appears to be a finely grained system of wealth redistribution. (For just two of many examples, consider all the rules on what constitutes a “public performance,” or the rules regarding compensation via the CRB (nee CARP) for compulsorily licensed performances (at Though I will admit that Michael Carrier’s work comparing property and IP has pushed me back toward seeing the appropriateness of the “property” analogy.

  3. A.J. Sutter says:

    This exercise is a wonderful illustration of the fallacy of “information flows”. What do these graphs mean?

    Google Labs’s explanation (see supplemental online material to the Science Express paper) distinguishes three types of data: the number of times an ‘n-gram’ appeared, the number of pages on which it appeared, and the number of books in which it appeared. For example, Chisum’s multi-volume patent treatise in and of itself might skew the results, if the right category isn’t viewed.

    Which type is represented in the percentages in each of the graphs above?

    The current version of the Google corpus doesn’t include part-of-speech tagging (SOM@25), or control for plurals and other grammatical declension and conjugation. So a search for “patent” doesn’t include “patents” (try it) and doesn’t distinguish the use of “patent” as an adjective or verb. Again, this could affect the metric, depending on the category. The Google corpus is also different from other linguistic corpora in that collocations aren’t visible. I.e, the context of the word’s usage can’t be shown. The mere fact that an “n-gram” appears in the historical record on a certain number of occasions doesn’t tell us whether or how its meanings or connotations have changed. This, in turn, ought to make us cautious in our interpretation of the graphs.

    By the way, the paper in Science Express makes clear that the tool relates only to about 1/3 of Google’s “vast repository of digitized books.”

    A prime motivation for this tool seems to be that Google wanted to rush into print with a fancy new word, “culturomics,” which “extends the boundaries of rigorous quantitative inquiry to a wide array of new phenomena spanning the social sciences and the humanities.” No more of that touchy-feely, qualitative stuff. (Sc., All your base are belong to us.)

    Does beguilement by pretty graphs whose meaning isn’t clear, or whose ambiguity is glossed over, really constitute knowledge? (Sorry, awkward question.) If this sort of uncritical usage of the graphs is meant to exemplify “access to knowledge,” it’s not at all encouraging about the referent of that 3-gram.