Iraq party lists (yes, again)

In the past, we have had discussions here about the type of lists used in Iraqi elections. All are in agreement that the elections of 2005 were by closed list, and that more recent elections were not. However, there has been some uncertainty about just which form of non-closed lists have been used.

In various previous discussions (click “Iraq” in the “planted in” line to see them), some of my valued commenters have linked to items from the Iraqi electoral commission that purport to show that the 2009 provincial elections were by flexible list, and that this year’s national assembly elections were by open lists.* Unfortunately, all those links now simply take one to the main Arabic page of the commission (and clicking there on the English link also does not seem to allow one to find archived articles).

I wonder if anyone saved these original articles, or has any other reliable sources** that clearly indicate the list format in these elections.

* The distinction that I am making is that under flexible lists, “preference” votes cast for candidates on a party’s list affect the order of election only for those candidates who receive some legally stipulated quota of preference votes. Otherwise, a pre-set party list order prevails. Under open lists, on the other hand, preference votes alone determine the order in which candidates are elected from the list, there being no pre-set list order with any affect on candidate ranking.

**In my experience, many writers, even by political scientists, will say “open” even when the lists in question in some country actually are flexible. (For that matter, sometimes that will refer to flexible lists as though they are closed. Flexible lists are kind of an orphan category, notwithstanding that they are used in so many European PR systems!)

10 thoughts on “Iraq party lists (yes, again)

  1. There are two things going on here, yes? One is a lack of detail about specific individual systems (Iraq’s in particular); the other the absence of a proper taxonomy of list systems.

    I’m thinking, for example, of Norway’s system, which, if I understand it aright, is a classic closed list system—except that anybody can put forward a list of their own by petition. That’s a fundamental sort of openness, at least in the ordinary sense of “open”, but we don’t have a word for it. (Well, maybe we do; I didn’t know what “flexible” meant in this context either.)


  2. Jonathan, I defined flexible in the footnote for exactly that reason. I agree that Norway is hard to classify, but most cases that are not fully closed or fully open are not ambiguous, using the criteria in my footnote. Norway is, for the reason you mention: the ability of voters to remake a list.

    I know some scholars are not comfortable calling Norway a closed-list system, but I am. It does not have a mechanism for a voter to vote for one or more candidates within a party-presented list. Cases like Belgium and Slovakia are straightforward here: they look like open lists in terms of ballot structure and how a vote is cast. But flexible and open lists differ in how the preference votes affect the final order.

    Still hoping someone can answer the question on Iraq…


  3. The more I read the comment by WIlf Day, referencing an item by Reidar Visser, on the provincial electoral law, the less I am persuaded that this was a flexible list.

    Another comment by Wilf on the national electoral law suggests open lists were more clearly the system adopted for this year’s elections. (But this comment references the items I alluded to above for which the links no longer direct one to a specific article.)

    Could it be that both elections (2009 provincial and 2010 national) were open lists: preference votes alone determine the order of election from the list?


  4. For the national election of 2010, it is quite clearly open list, adjusted to ensure the women’s representation quota is reached.

    Within each winning list, candidates are ranked by the number of their individual votes, from highest to lowest. In principle, this ranking will determine which candidates will be awarded a seat. However, this initial ranking will be adjusted if necessary to fulfill women’s quota.


    The governorates with the lowest share of elected women will be identified. These will need to increase the number of winning female candidates until the number in the Council of Representatives reaches 82. Within each of the governorates so identified, the winning lists with the lowest share of elected women will be identified, and these will need to increase the number of women, until the required number is reached for the governorate. Within each list so identified, the list of candidates will be re-ranked so that women candidates move up the list to winning positions to replace male candidates, until the required number is reached for that list.

    See the second fact sheet at And thanks to Wilf Day (in an e-mail) for pointing this out.

    Unfortunately, the page for “fact sheets” on the 2009 provincial elections is empty, so this still does not provide a definitive answer to the question of whether the list type for those elections was fully open or not.


  5. An organization called has a fact sheet for the provincial elections that says those were also held by open list.

    The 2009 Provincial Elections will take place using an open list, proportional representation system. At the polls, Iraqis will select both a party, and individuals within that party. Representation on the provincial council will be based on the percentage of votes garnered by each party, with the membership in the council subsequently being given to the candidates with the highest percentage of the vote, rather than their position on a party?generated list. [Footnotes suppressed]

    So mystery solved, apparently.


  6. Can’t we just use “fixed” “semi-fixed”, and “unfixed”, with the middle category being further measured by what percentage of MPs are elected solely or mainly because of their list position (ie, they were one of the SEATSWON highest-ranked candidates but not one of the SEATSWON highest-polling candidates).


  7. I’m making a somewhat smaller (and in this context somewhat digressive) point: we’ve assigned the terms “open”, “closed” and (apparently) “flexible” certain meanings in the context of list-pr elections.

    Norway’s system is certainly not “open” in this sense; rather it’s in its own way open in the ordinary-English sense of “open”. If open-flexible-closed is a continuum of sorts, Norway’s variation goes off in another direction (and a useful one, I think).

    I assume that there are other dimensions that are relevant, such as the means by which candidates are ordered on closed lists, or the details of the mechanism by which seating order is determined in open and flexible systems.

    And that goes to the desirability of a taxonomy by which we can capture the similarities and differences of these various systems with a mutually intelligible set of terms. Tom Round’s dimension of fixedness is perhaps a step in that direction, but only a step.


  8. “Flexible” is a poor term, given that so many of the systems so-called require a high quota of preference votes to alter the order, and hence are not very flexible in practice. Clearly it is a large residual category to describe what is a continuum.

    Open and closed, on the other hand, describe clear end-points of that continuum. As for how the lists are ordered, under closed lists it is entirely up the party (using whatever rules it might use), while under open lists it is up to the voters (giving their preference votes across the set of candidates the party has nominated).

    While better names could be devised for the intermediate category (semi-open is sometimes used, as semi-closed), I would be reluctant to rename the systems at the endpoints. We have enough definitional clutter as it is!


  9. By all means, “open” and “closed” should be the endpoints of one particular continuum. The problem, though, is that while there are indeed intermediate points on that continuum (whether we call them flexible or semi-something), there are also other dimensions (Norway, for example) that aren’t adequately described by simply placing them somewhere on the open-closed line.


  10. For national elections, Norway has a system that is open in theory, though not in practise:

    “For the electors’ changes on the ballot papers to have any influence on the choice of person more than half the party’s electors must have made changes in the case of the same candidate. This means that it takes a very great deal for the electors’ changes to have any significance. In practice this never happens.”

    In local elections, however, it is much easier for changes to election lists to be taken into account. While I don’t think there has ever been a case, at least not in modern times, of someone being elected to the parliament because voters remade the list, this happens all the time in local elections.

    I takes slightly more than a petition to put forward a list. In parliamentary elections, registered political parties can put forward lists in any constituency, if they received at least 5000 votes in the whole country, or 500 votes in a single constituency, at the last parliamentary election. Others can put forward a list if it is signed by at least 500 people with voting rights in the constituency. The smallest constituency, Finnmark, currently has 72 981 inhabitants. Don’t know whether that makes it easy to put forward a list or not, compared to other countries.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.