…so I like looking up data 
Originally I posted NSF data from (1988 to 2013)
http://www.nsf.gov/statistics/sed/2013/data/tab31.pdf
It showed time-to-degree since bachelors as going from 10.6 to 9 years. This motivated me to look further back.
I found one NSF report going back to 1978 (page 3)
http://www.nsf.gov/statistics/infbrief/nsf06312/nsf06312.pdf
In 1978 the time to degree was 9 years…just like it is today. Interestingly there was a sharp increase from 1978 to late 1990s to up to 10.8 years. This was followed by a sharp decrease in the 2000’s and 2010’s back to 9 years.
Going back even further I found a document that went back to 1920 (page 48 in adobe reader or page 36 in the document itself.
http://www.nsf.gov/statistics/nsf06319/pdf/nsf06319.pdf
In 1975 (40 years ago) the time to degree was ~9 years…just like today. In fact the time to degree varied from 8 to 9 years between 1942 and 1975 (e.g. it was also ~9 years in 1945). You would literally need to go back to 1940, before the US entered WWII, to see a PhD take less than 8 years across all fields.
As for this comment by Leonard Cassuto
It’s nonsense, and it should not have been published in the Washingtonpost.