« What the Binary-Scalar Distinction Misses | Main | Isn't There a Tradeoff Between Liberty And Security? »

Tuesday, November 04, 2014

Election law as contextual: a universal truth? (And, happy election day to U.S. readers!)

I am grateful to Dan Markel for this chance to spend another month in conversation at Prawfsblawg.  As with my last go-around, my focus is on U.S. election law.  This time, however, I get to talk about election laws on an election day. 

 When the voting and vote counting unfold, we’re bound to see election laws and administrative practices in the news.  Even if the odds-makers are proven correct in their forecast of an election day that is characterized by relatively low voter turn-out and relatively few close contests, there will be questions or controversies about the effects of heightened voter identification requirements, the counting of provisional ballots, the scheduling and ballot design for a gubernatorial run-off, and the like.  Those of us who follow politics have come to instinctively associate some of these contested laws and practices with a particular effect (a tendency to expand or narrow the electorate), and with a particular political valence (a tendency to disenfranchise or dilute the votes of one or another party or racial or socioeconomic group).  

Of course, election rules, such as the new voter identification requirements in Texas, will, at times have their strongest bite in the lives of individuals (see, e.g., Eric Kennie’s story at  http://www.theguardian.com/us-news/2014/oct/27/texas-vote-id-proof-certificate-minority-law). But politicos and scholars usually train their attention more on election rules as they might tip a contest for a particular candidate or party.  To be sure, different political camps tend to have different empirical and normative premises about election rules’ operations.  Voter i.d. requirements are about culling the poor, the disabled, and racial minorities from the electorate.  They are a procedural tool for disenfranchising eligible voters.  Or, no, these requirements are about screening out fraud and low-information voters.  They are about protecting the eligible and informed voters from vote dilution.  All sides, however, can instinctively agree on a rule’s expected effect and valence:  Strict voter i.d. rules contract rather than expand the electorate, and they can be expected to do so to the benefit of Republicans.

 I now want to take many steps back from the immediacy of these voter i.d. rules and today’s election.  (It’s not like you have any election results to follow!)  I want to consider whether perceived regularities in the consequences of elections laws (large and small) may hold true across many different contexts.  

 Political scientists (one of my tribes) have often assumed that the answer is “yes”, and they have precisely defined their scholarly enterprise to be a search for the generalizations that will not be context-bound.  The successes of this research program have been real.  We have learned that election rules can exhibit regularities, sometimes ones that operate behind the backs of the political actors.  A particularly successful example is Duverger’s Law which states that legislative elections by single-member-district and ‘first-past-the-post’ rules (such as in the U.S., Canada, and Great Britain) are correlated with two-party systems while proportional-representation rules are correlated with multiparty systems.  

 This generalization is powerfully universal.  Except when it isn’t.  Many times, political scientists have found the need to qualify it.  It fails to hold true in a country where there is no widely shared information or expectations about the different parties’ electoral prospects, or in a political culture where voters do not mind ‘wasting’ their votes on a third-party candidate who can’t win (Powell 2013).  It fails to hold true in a federal system at the national level if the national parties are really sectional parties (Chhibber and Kollman 2004.)  And so on.

 If even Duverger’s Law is highly context-bound, then we may suspect that there are few, if any, (non-trivial) regularities in the consequences of election rules that are not similarly context-bound.  And in fact, G. Bingham Powell has used this example to make a (to me) compelling case that the proper study of the scientific ‘laws’ of election law can’t be (or, at least, it can’t be restricted to) a search for big universals.  Even when generalizations are prized over local knowledge, election laws need to be studied closer to the ground in order to unearth the local and temporal conditions that may limit an otherwise robust pattern, or that may set in motion a new one.  

 Duverger himself recognized that the consequences of election rules are mediated by context, and he classified some of these contextual factors as (1) “the mechanical” (the interaction between votes and election rules if the latter are properly administered—conditions that may depend on the strength of a country’s tradition of rule of law and technical competence)  and (2) “the strategic” (the effects of citizen or elite anticipations of these mechanical operations).  

 We might think about recent voter identification laws in a similar fashion:  Under current conditions, heightened documentation requirements can be expected, at least at the margins, to disproportionately shave the vote totals for some Democratic-leaning constituencies.  This effect may seem almost mechanical.  Yet, as we have apparently witnessed in recent years, some election reforms that raise the costs of voting for particular classes of voters (such as proof of citizenship requirements, or cut-backs in early voting days like ‘Souls to the Polls”) can occasionally result in an increase in the vote totals through the mechanism of ‘backlash’ mobilization against the reality or perception that the reform was an intentional form of disenfranchisement.  (On such backlash, see, e.g., Rick Hasen’s Voting Wars).  My (perhaps, not so social-scientific) spin on this example:  human agency and innovation matter.

 Powell offers his insights about the contextual nature of election law for the sake of a positive research program into election laws’ consequences.  I, however, want to use these insights to conclude with two simple points that are more normative in nature.  

 First, as citizens or election reformers, the contextual nature of election rules means that we should be wary of categorical judgments about particular election rules.  Changes in the environment, human behavior, or the law's internal design may flip expected realities.  (Just as, at one time, the secret ballot served to free humble tenant voters from the pressure of their landlords, so at another time and place, it worked to disenfranchise the humble illiterate…)  Voter documentation requirements, for example—if they are the responsibility of government, and not voters themselves—may have an entirely different effect and valence than what we’ve come to expect in the U.S.  

 To judge from the experience in some countries at least, it seems possible that voter documentation can operate to expand, not contract, the electorate, and that it can operate without benefit to a particular party (other than the ‘partisan’ benefit that is likely to accrue from fully documenting an eligible electorate).  If this is right, then—yes, of course—government-controlled voter i.d. will run into other objections (such as those of the civil libertarians worried about runaway uses of national i.d.).  But the point stands that our political (politicized?) instincts about the natural effect and valence of voter id would no longer hold.

 Second, if the consequences of most or all election rules are highly context-bound—meaning that an election law that is benign in one context can be malign in the next—then the quality of our processes and institutions for evaluating and changing election rules may be far more important than the static quality of any particular election rule.  I’ll say more about this latter point at another time.

 Now back to the immediacy of election results and (perhaps) election administration debacles.


 Pradeep Chhibber and Ken Kollman, The Formation of National Party Systems: Federalism and Party Competition in Canada, Great Britain, India, and the United States. Princeton: Princeton University Press, 2004.

 Maurice Duverger, Political Parties: Their Organization and Activity in the Modern State. New York: John Wiley, 1954.

 Richard L. Hasen, The Voting Wars: From Florida 2000 to the Next Election Meltdown. New Haven: Yale, 2012.

 G. Bingham Powell, Jr., “Representation in Context: Election Laws and Ideological Congruence Between Citizens and  Governments,” Perspectives on Politics, Vol. 11/No. 1, March 2013.

Posted by Kirsten Nussbaumer on November 4, 2014 at 04:22 PM in Constitutional thoughts, Law and Politics, Legal Theory | Permalink


Post a comment