Why is vocabulary shrinking?

Vocabulary decline in adult non-Hispanic White Americans (controlled for years of education completed)



"Are Americans more intelligent than a few decades ago, or less intelligent?" So asks psychologist Jean Twenge in her introduction to a recent paper on vocabulary decline in Americans. The findings are disconcerting, to say the least:

We examined trends over time in vocabulary, a key component of verbal intelligence, in the nationally representative General Social Survey of U.S. adults (n=29,912). Participants answered multiple-choice questions about the definitions of 10 specific words. When controlled for educational attainment, the vocabulary of the average U.S. adult declined between the mid-1970s and the 2010s. Vocabulary declined across all levels of educational attainment (less than high school, high school or 2-year college graduate, bachelor's or graduate degree), with the largest declines among those with a bachelor's or graduate degree. (Twenge et al. 2019)

The last decline was especially large: more than half a standard deviation. In general, vocabulary test scores have fallen by 8.5%. Ethnic change doesn’t seem responsible, since non-Hispanic whites have had almost the same decline: 7.2%.

So what's going on? The authors considered the explanation they first raised: Americans have become less intelligent despite the increase in education.

First, Americans' vocabularies might be shrinking despite the increase in education. This is plausible given the steep decline in the amount of time high school students spend reading [...] and the decline in SAT verbal scores over time [...]. This explanation could account for the narrowing of abilities between those without high school educations and those with college educations. The difference in vocabulary by education was approximately 3.4 correct answers in 1974-79 but dropped to 2.9 correct answers by 2010-16. However, this explanation would not account for the decline in performance in all educational groups. (Twenge et al. 2019)

Uh, why not? The last sentence makes sense if the explanation is simply that postsecondary education has become less effective. But what if vocabulary has declined because the capacity for learning words and retaining them has also declined? The cause may be genetic. Can we at least ask that question?


Lower admission standards? Mismatch between cause and effect

The authors then consider another explanation: because college admission standards have been lowered, people of lower ability have been going on to postsecondary education in larger numbers; those who don't are increasingly the least able.

If education does not improve vocabulary, but educational attainment increases, those with higher ability will be increasingly selected into the higher education groups, leaving those with lower ability in the lowest educational attainment groups. Thus, the no high school degree group will be left with those of lowest ability, and the college graduate group will have absorbed more with only moderate ability. (Twenge et al. 2019)

That explanation is popular, but it doesn’t really match the findings. The vocabulary decline was steepest during the late 1970s and early 1980s. It then levelled off. A second decline may have begun in 2008, but it’s still too early to say (see Figure 1 reproduced above). Most of the decline doesn't correspond to any previous change in college enrollment by recent high school graduates. The enrollment rate rose slowly from 45.7% in 1959 to 49.4% in 1980. It began to grow faster only in the mid-1980s, breaking through the 60% level in 1991 and the 70% level in 2009 (Bureau of Labor Statistics 2010).

So the alleged cause doesn’t match the presumed effect. The steep increase in college enrollment from the mid-1980s onward could not have caused the steep vocabulary decline during the late 1970s and early 1980s. Keep in mind that most of the GSS respondents had completed their education some years earlier, almost ten years earlier on average. So the average respondent in the late 1970s had to meet college admission standards that existed in the late 1960s.


Most of the decline has been among early boomers

Because the GSS was first administered in 1974, we don't know when the steep vocabulary decline began. But we do know when it ended: in the mid-1980s, among respondents who were born on average thirty years earlier. A genetic cause would imply a rapid deterioration in the gene pool from 1945 to 1955 and a slower deterioration thereafter. I have no idea what that cause could be.

If we're looking for a cultural cause, it would have acted most strongly on the same cohort of "early boomers." Perhaps it was their increasing exposure to TV and their decreasing exposure to high literature. Those cultural changes were already a fait accompli for "late boomers," who experienced a more gradual dumbing down of vocabulary on TV and in print. The post-2008 vocabulary decline, if it’s real, might reflect the growing importance of iPhone texting since the late 2000s.

That cultural explanation has some support from the data and is favorably mentioned by the authors. For one thing, comparison with the results of another test (WAIS) suggests that the decline has been mostly in passive vocabulary, i.e., the words we understand but don’t use spontaneously in speech (Twenge et al. 2019). We’re less proficient in "bookish" language:

Perhaps American culture became less intellectual, either because of or in response to a lowering of verbal ability among those who read books. Authors aim to sell more copies of their books, and thus may adjust their vocabulary level to the skills and preferences of a wider slice of the population. Or, perhaps authors lowered the vocabulary level of their books for some other reason such as an interest in getting out a message without linguistic complexity getting in the way. For example, the Bible has been revised repeatedly to make it more accessible with the King James Version, the most complex and lyrical English language version, being succeeded by the simpler New International Version, Living Bible, and New Revised Standard Version. (Twenge et al. 2019)

The last point rings true. When I was studying Shakespeare in high school my mother could explain words I had trouble understanding. She had never gone beyond Grade 10, but she could read the Bible in the King James Version, as well as a lot of high-brow literature. This was true for many ordinary adults in the 1970s. Today, regular reading of the Bible is unusual and almost always confined to modern English versions.


Conclusion

Yes, college has become a less interesting place for learning vocabulary, and for learning in general. Yes, a big reason is the growing number of students who don’t really belong there, and the consequent lowering of standards. Yes, America’s cultural and linguistic mix is changing, and for that reason alone the average American would have a smaller English vocabulary.

Nonetheless, those factors fail to explain why non-Hispanic white Americans know fewer words today than they did a half-century ago, especially in their passive vocabulary. Something else is going on, and it seems to be a shift away from high literature and toward simpler audiovisual media: TV, video, text messaging …


References

Bureau of Labor Statistics. (2010). College enrollment up among 2009 high school grads. TED: The Economics Daily. April 28
https://www.bls.gov/opub/ted/2010/ted_20100428.htm?view_full

Twenge, J.M., W.K. Campbell, and R.A. Sherman. (2019). Declines in vocabulary among American adults within levels of educational attainment, 1974-2016. Intelligence 76: 101377
https://www.gwern.net/docs/iq/2019-twenge.pdf