Electoral Laws and their Political Consequences, eds. Bernard Grofman and Arend Lijphart

When I finished the books on my shelves acquired in 2016, I knew I had mislaid several about elections and boundaries, and to my delight I located those a couple of weeks ago and am now going through them.

This is a collection of no less than 18 essays on electoral laws by the top-ranked political scientists of the day (early 1980s, published in 1986). All but one of the authors is male; all but one of the authors is white. The third chapter is a reflection by Maurice Duverger on “Duverger’s law”, first proposed by him forty years earlier, which pointed out (though in terms that are disputed) that majoritarian electoral systems tend to go with strong two-party political systems, whereas proportional electoral systems tend to go with multi-party political systems. The second paragraph of the third chapter, which includes a quote in the middle, is:

Certain errors of interpretation have resulted from my own tentative and imprecise formulations. An example is the alleged difference between “Duverger’s law” on the plurality rule’s tendency to create and maintain two-party systems and the “hypotheses” concerning the tendency toward multipartism of proportional representation and the two-ballot majority system, Riker’s (1982a; see also, chapter 1) analysis of this distinction is the most recent instance. In 1951, I did say in Political Parties that the former was “the closest to a sociological law among all the generalizations suggested in this book,” but this remark did not have the significance that was later attributed to it. It simply reflected my cautious attitude which was a reaction to the criticisms of the propositions that I first stated in 1945 at a conference at the University of Bordeaux (Duverger, 1946a, 1946b), where I presented the con-sequences of the three electoral systems as a “threefold sociological law.” I already discarded this expression in the paper presented at the 1950 Congress of the International Political Science Association which merely mentioned “three formulas” (Duverger et al., 1950, p. 13). But later I used it again in the first edition of my handbook Droit Constitutionel et Institutions Politiques which propounds “three sociological laws defining merely basic tendencies that interact with national and social factors” in terms which have hardly changed since then.

(1) Proportional representation tends to lead to the formation of many independent parties, . . . (2) the two-ballot majority system tends to lead to the formation of many parties that are allied with each other, . . . (3) the plurality rule tends to produce a two-party system (Duverger, 1955, p. 113).

In the more recent editions, the second law is formulated as follows: “The two-ballot majority system tends to produce multipartism tempered by alliances.”

The thing about Duverger’s ‘law’ is that it’s obviously true except when it isn’t. Majoritarian systems don’t always lead to concentration around two alternatives – Canada and India, and even the UK to an extent, have seen the two-party system rise and fall and rise again. And nobody ever mentions Malta, which despite having a proportional election system very similar to Ireland’s, has a rigid binary political divide – the Labour Party and the Nationalist Party between them have won literally every parliamentary seat since independence sixty years ago.

In general I found that the arguments here were largely in issues that I considered in my 20s, soon after the book was published, and a lot of it seemed very old-fashioned. There is a lot more experience of democratic systems now than there was in 1985, given that we have had an end to single-party politics in most of Eastern Europe, and the debate between whether proportional or majoritarian systems are better is basically over, after the failure of the 2011 referendum in the UK and the Trudeau government’s decision not to proceed with reforms in Canada in 2015. There’s also a lot of discussion of peculiar US electoral practices that the rest of the world is unlikely to copy.

Still, there were a couple of chapters that really stood out. One, by Gordon E. Baker, looked at the revolution in reapportionment in the US, and made the point that the shift to insisting on numerically equal populations in each state’s Congressional districts, plus various other contradictory court findings, has actually made it more difficult rather than easier to draw fair boundaries. Forty years on, I fear that this hasn’t changed. And Peter Mair has a lucid paper on gerrymandering in Ireland with multi-member constituencies and the Single Transferable Vote.

However, there’s also a crashingly unreadable review of (then) recent literature compiled by Taylor, Gudgin and Johnston (who were all capable of much better); and the most annoying thing about my 2003 reprint is that the OCR’ed typesetting has been poorly edited and the placement of the letter ‘f’ is irritatingly off-centre in the words “of” and “if”.

As I said, this is one of a cache of election-related books that I’m now going through. Next up is Comparing Democracies, edited by Lawrence LeDuc, Richard G. Niemi and Pippa Norris. But you can get this one here.

Version 1.0.0