I have been doing some work in the LSE where I came across a really interesting early reference to semi-presidentialism.
The reference is in Encyclopaedia of the Social Sciences, Volume 7 (Editor-in-chief, Edwin Robert Anderson Seligman; Associate editor, Alvin Saunders Johnson), Macmillan Co., 1932, pp. 75-81. The entry is on ‘Succession States’ (meaning successors to the Austro-Hungarian and Ottoman Empires) and it is written by Robert Braun.
Talking about the 1929 constitutional reform in Austria, the relevant part of the text reads “from a purely parliamentary republic Austria has been changed into a semipresidential republic”.
This is a really interesting reference, to me anyway, for three reasons.
Firstly, it is a very early reference that I had not come across before. So far, I have found two 19th century references, a cluster of references in 1919-1923 relating to David Lloyd George, and a reference relating to the US president in 1930. So, this 1932 reference is a very early one.
Secondly, it is the very first reference I have found that uses the term ‘semi-presidential’ to refer to a constitutional arrangement. Prior to this time and for most references until the early 1950s, the term is used to refer to the situation where a prime minister is more powerful than normal or where a president is less powerful than usual. However, the reference here refers explicitly to Austria’s 1929 constitutional amendment, which introduced the direct election of the president into the previously parliamentary system. The earliest constitutional reference of this sort that I had found previously dated to 1938. So, again, this is an early reference.
Thirdly, it uses the term in relation to Austria. I class Austria as semi-presidential from the constitutional reform in 1929 to 1933 when the constitution was suspended. However, most Austrian observers do not, even the ones who class Austria as semi-presidential after 1945 under a very similar constitutional arrangement. The reasons, I think, why most people do not class Austria as semi-presidential from 1929-1933 is because the constitution was suspended before there was the opportunity for a presidential election and because the incumbent president was a fairly minor figure, certainly the prime minister was the key political actor. So, to see the term ‘semi-presidential’ used contemporaneously in relation to the Austrian First Republic is very unexpected.
Presumably, old references will re-emerge as more books become digitised. However, this reference will remain, I think, a particularly interesting one.