When discussing a report last year on devolution and accountability in Vietnam , we faced the challenge of distinguishing accountability from responsibility since they are so similar when translated into Vietnamese. Similarly, Bank documents often find it necessary to explain the difference between (bad) governance and corruption. Clearly, we think these differences are important or we wouldn’t bother explaining them.
Although we might not be able to cleanly track how people are using these terms, we can track how often they are used in print. The Google Books Ngram Viewer gives you a way to see how words rise and fall in prominence over time by mining a database of words and terms from the millions of books that they have scanned. Why not use the tool to examine the use of various governance terms over time? Are these terms complements, rising and falling in popularity together, or substitutes, with one supplanting the other? Are they at the dawn or the dusk of their popularity, or holding steady at high noon?
I’d like to say I came across the idea after some serious reading about “culturomics” in Science, but the truth is I learned of the tool from a column by Gene Weingarten. Whether the purpose is serious scientific research (Adam Smith vs. Karl Marx) or good fun (Weingarten’s Harry Potter vs. Huckleberry Finn), the tool is hard to resist. And the results can be both intellectually interesting and fun at the same time.
Caveats first: The results naturally have some false positives, such as when words have multiple meanings (e.g., institutions). And the results may also turn up false negatives due to OCR errors (e.g., 1nstitutions would not be picked up). But choose your words carefully, and you can have great fun.
So grab some grains of salt, dive into the Ngram river and get a feel for the ebb and flow of some key governance terms…
Two practically synonymous terms that we use frequently are administrative corruption and bureaucratic corruption. The former’s dominance of the latter was reversed for a time from about 1980-2000. Add in state capture and the picture becomes more interesting, especially as that term came to prominence in 1999-2000 with the publication of the Bank’s Seize the State, Seize the Day and the first Anticorruption in Transition, and EBRD’s Transition Report 1999.*
All of these terms have been and continue to be dominated by political corruption.
The considerable overlap between the concepts of institutions and governance also begs analysis. Institutions are defined by Douglass North as the “humanly devised constraints that structure human interaction”, while the Bank’s GAC Strategy defines governance as “the manner in which public officials and institutions acquire and exercise the authority to shape public policy and provide public goods and services.” Although these are not identical, they certainly cover a lot of the same ground. Given that institutions is a word with several meanings in the English language, it absolutely dominates the term governance. And although the increasing use of governance reform mostly coincides with a decline in the use of institutional reform, references to good governance took off long before the decline in institutional reform.
Turning to some key tools in the governance toolkit, accountability passed transparency around 1970, although both have been on the upswing through around 2000.
Returning to the dichotomies discussed at the outset of this blog, corruption dominates governance through most of (google books’) history, but governance took the lead some time after 2000—the two terms come across as complements, rather than substitutes. Alas, accountability’s rise seems to have coincided with a decline in responsibility.
As the tool covers only the period from 1500 to 2008 (only!), it is less well suited for more recently coined terms. Although Mancur Olson’s stationary bandits first made an entrance in the early 1990s, the term’s popularity took a decade to cement. In contrast, while North et al’s limited access orders already calls up more than 18,000 results when googled, the term does not yet register as an Ngram. (A term does not register unless it appears at least 40 times in the sampled books.) We’ll have to check back in a few years. Until then, we can continue to amuse and enlighten ourselves with the ebb and flow of more established terms.**
• peace still dominates prosperity.
• efficiency surpassed equity late in the 19th century, although equity seems to be fighting a comeback since about 1970.
• economic growth apparently took off as a term in the English language only after about 1950, while terms like poverty reduction and poverty alleviation did not really take off until the 1980s; economic growth still dominates the two poverty terms combined.
• human rights diverged significantly from property rights around the 1970s, although both terms increased in prominence in the 1980s and 1990s.
• love kicks hate’s butt all through time.
*These charts were all done with 3-year smoothing which is the default for the n-gram viewer, and all cover the period 1800-2008 using the basic English corpus. Yes, such fun can be had in a range of languages. As the fonts generated by the tool are microscopic when shrunk to bloggable size, I’ve annotated the charts for easier viewing.
**Since you’ve read this far, I will end the suspense: Marx caught up to Smith around 1940 and took the lead in the 1970s, with Smith pulling neck and neck about a decade ago. Potter passed Finn around 2001. In the tag team, Marx and Smith have no trouble with Finn and Potter.