politics : culture : economics

Why Turkey’s new constitution will be an opportunity missed

In A new constitution on October 5, 2010 at 11:02 am

In my last post, I quoted Juan Linz and Al Stepan outlining the kind of conditions that need to prevail in order for a country to engage in a process of “free and consensual constitution-making”. I think the chances of Turkey meeting (or even approaching) this benchmark are non-existent or thereabouts.

That Turkey is soon to embark upon the process of writing a new constitution is, or ought to be, a remarkable development. This is not how these things tend to happen. New constitutions are generally written ‘hot’, following a crisis such as revolution or war. This lends the constitution-making process an inevitable air of significance (albeit one that the resulting documents and their implementation don’t always manage to live up to).

Turkey, by contrast, is coming to its constitution-making cold. Yes, the country is riven by heated differences. And yes, its democracy remains unconsolidated. But it has managed to muddle through, for decades now, with a series of demonstrably problematic constitutions. And this muddling through appears to have been habit-forming. Because when the party leaders speak of the constitution-making that now lies ahead, there is no sense of a country on the cusp of something new, as surely there ought to be. (In theory, we are talking about Turkey’s first properly democratic constitution here. In theory, this is big deal stuff.) Rather, drafting the new constitution appears, for government and opposition alike, to be just another item to squeeze into a crowded legislative calendar.


Of course, writing a constitution isn’t legislative business in any ordinary sense. But before looking at the relationship between constitutions and legislatures a bit more closely, I want to touch on the question of having a mandate for constitution-making.

I do so in the context of the recent suggestions, from both president Abdullah Gül and CHP leader Kemal Kılıçdaroğlu, that a new constitution might be drafted before the general election that’s due in the summer of 2011.This proposal doesn’t seem to have made any headway. But the fact that it was mooted at all speaks volumes about the way in which this country’s political class views the public’s role in the constitution-making process.

At this stage, the current parliament has no business writing a new constitution. Arguably when elected it had the authority to do so, but that authority has by now been exhausted, following the 2008 headscarf amendment crisis and the recent passage of the 12 September referendum. There are only so many times that parliament can credibly return to the constitutional well without seeking to renew its mandate from the electorate.

This isn’t just an dry point of principle. From the opposition’s point of view, it’s one of raw politics too. The 2007 elections took place in depressingly exceptional circumstances, and the votes that were cast reflected those conditions. If Mr Kılıçdaroğlu thinks that the framing of a new constitution should be entrusted to a legislature elected on the basis of a 47 per cent vote for the governing AKP, then more fool he.


Turkey is not alone in turning to its legislature to craft a new constitution. Most new constitutions are written in this way. That doesn’t make it the most desirable way of doing things. Jon Elster, who has given these questions more attention than most, suggests that the use of legislatures rather than specially convened constituent assemblies generally comes down to a lazy choice of the path of least resistance. “When a legislature is in place, why not use it?”*

A potent argument against using a legislature as a forum for constitution-making relates to the compositional differences that ideally ought to exist between the two. Elster:

In modern societies, an important function of the legislature is to provide the basis for an effective government. It is widely agreed that for that purpose, one wants an electoral system that limits the number of parties represented in the assembly. Majority voting or proportional voting with a high threshold are generally agreed to promote this end. To the extent, however, that the task of a constituent assembly is to promote free and unconstrained deliberation among all concerned parties, with a large variety of interests being represented, proportional voting with no threshold or a low threshold is more desirable.

Turkey, with its ten per cent electoral threshold, conforms more than most countries to the pattern of a restrictively composed legislature. In the current parliament, the effect of the threshold is that 18 per cent of 2007’s voters are unrepresented. Even this is a marked improvement on the results of the 2002 elections, when a shocking 46 per cent of votes were eliminated by the threshold.

The undemocratic nature of the threshold is now widely recognised in Turkey, and its removal for elections to future legislatures is likely to feature prominently in debates about the new constitution. That being so, its retention as a feature in the more fundamental exercise of writing a new constitution should raise eyebrows.


What is the alternative to relying on standard operating procedures in the legislature to draft Turkey’s new constitution? In an ideal world, properly proportional elections would be used to create a body tasked solely with working on the draft of a new constitution. This body would exist concurrently with the legislature, which would continue to fulfil its day-to-day role in the running of the state.

Such a constituent assembly would have an unambiguous and unpolluted mandate to draw up a new constitution. It would also, according to Bruce Ackerman, impose more powerful democratising incentives on participants than would operating within the legislature. Reflecting on the framing of America’s constitution, Ackerman states that:

By establishing a convention apart from ordinary organs of government, the revolutionaries did more than isolate the problem of constitutional order from the many short-term issues that bulked large on the political agenda. The very creation of the separate forum created new and powerful political incentives for a successful conclusion of the experiment in constitutional formulation.

These incentives operated in a very personal way that all politicians can understand: quite simply, the revolutionary leaders risked making fools of themselves if the convention dissolved without achieving an acceptable constitutional text. … They were self-consciously creating an incentive structure that maximised the chances of success by imposing new penalties for failure.**


To repeat what I said at the outset, I think the chances of Turkey implementing anything like this kind of arrangement are as close to zero as matters. Next year’s elections won’t be properly proportional and so won’t ensure the broadest possible degree of representation in constitutional debates. And the elections will be to a standard legislature rather than to a separate constituent assembly.

Dishearteningly, the absence of these features won’t result from reasoned discussion, nor even from a conscious decision to take the line of least resistance. There is simply no sense that the mechanics of constitution-making are even on the radar as something requiring political attention.

Ackerman notes that “a piece of paper calling itself a constitution can be many things: an empty ideological gesture, a narrowly legalistic document, or a profound act of political self-definition.” Turkey desperately needs a democratic constitution of the third variety. But in all likelihood it it will end up with something much more closely resembling the first. I would like to be proved wrong, but ultimately the democratic interests of the country’s politicians (both in the sense of their curiosity and their incentives) appear to be much too shallow to deliver us any more than that.

* Jon Elster, “Legislatures as Constituent Assemblies” in The Least Examined Branch: The Role of Legislatures in the Constitutional State.

** Bruce Ackerman, The Future of Liberal Revolution.

  1. Excellent post, to the point. bravo!

  2. In fact a major amendment did get passed in 2007 and gave us a time bomb. We’ll have a crisis about presidency either in 2012 or in 2014 — we don’t even know when.

    As for representation, the threshold isn’t the only problem. Representation is skewed in favor of smaller provinces. There can be an almost 4-to-1 difference in the weights of peoples’ votes based on where they vote. I don’t understand why that’s not talked about. (Try to find out how many votes it takes per MP in Istanbul and a small province to see this. I looked into this in 2007, and my back-of-envelope notes show about 89k and 21k actual votes cast per MP in Istanbul’s first district and Bayburt, respectively.)

    And, of course, it is clear that the MPs are appointed rather than elected. I don’t know who in the parliament is actually elected other than the party leaders and people like Kamer Genc.

    Those problems can be fixed without a new constitution, but they are not. This should tell us something about how sincere the parties are about many of those high-sounding concepts.

  3. Bulent,

    I agree completely with your point about the variations in the representative-to-voter ratio across the country. (Anyone looking for them can find a breakdown of the 2007 results by province here: http://psephos.adam-carr.net/countries/t/turkey/turkey20072.txt.) Not that it’s of any direct relevance to the Turkish case, but this is something which the Irish system deals with in the constitution:


    ARTICLE 16


    1° Dáil Éireann [Ireland’s lower house of parliament] shall be composed of members who represent constituencies determined by law.

    2° The number of members shall from time to time be fixed by law, but the total number of members of Dáil Éireann shall not be fixed at less than one member for each thirty thousand of the population, or at more than one member for each twenty thousand of the population.

    3° The ratio between the number of members to be elected at any time for each constituency and the population of each constituency, as ascertained at the last preceding census, shall, so far as it is practicable, be the same throughout the country.

    4° The Oireachtas shall revise the constituencies at least once in every twelve years, with due regard to changes in distribution of the population, but any alterations in the constituencies shall not take effect during the life of Dáil Éireann sitting when such revision is made.

    5° The members shall be elected on the system of proportional representation by means of the single transferable vote.


    I also agree with you on the closed-list system for appointing/electing MPs. There’s no surprise at all that Turkey’s party elites in successive parliaments haven’t revisited the closed list system given how much power it leaves in their hands. The (naive) hope would have been that the process of drafting a new constitution might have shaken things up and created an opening in which these kinds of issue might have received an airing, given the novelty/gravity of the process. But it’s pretty clear that’s not going to happen.

  4. A couple of further links in relation to the skewed weighting of parliamentary representation noted by Bulent above. The relevant electoral law can be found at these URLs:

    * TÜRKÇE: http://bit.ly/cXqKFR [html]
    * ENGLISH: http://bit.ly/943CUT [pdf]

    Article 4 concerns the number of deputies in each province. The first paragraph seems to be the element that skews the representation. Every province is guaranteed at least one representative, regardless of population. To take the two provinces mentioned by Bulent above, this guarantee is much more valuable to the 53,000 registered voters in Bayburt than it is to the 7,400,000 registered voters of the Istanbul province.

  5. Belge.net has all the data from 54 onwards:


    The election system here has been changed many times. We even had something called ‘milli bakiye’ that gave rise to a rather accurate proportional representation in ’65 (changed in ’69).

    I found a list here by quick googling: http://www.anayasa.gov.tr/files/pdf/anayasa_yargisi/anyarg23/tuncer.pdf (page 3)

  6. Bulent,

    I think one of the most striking bits of electoral change here was law 4774 from December 2002, facilitating Erdoğan’s election. Given that the last paragraph of Article 67 was added to the constitution in 2001 to prevent governments messing with electoral law for their own short-term ends, it’s quite something to see an amendment passed just a year later which suspends that paragraph for the duration of a single parliament. (This goes back to your initial point about the sincerity of the parties on some of the high-minded principles to which they pay lip service.)

  7. Hahaha, yeah, they did that together, CHP and AKP. The president at the time — I think rightly — vetoed it (once) on the grounds that the parliament cannot pass laws favoring a specific individual. I dug up the text: http://arsiv.ntvmsnbc.com/news/193445.asp

    You can have more fun by looking into things like at whose expense RTE finally got into the parliament, who he ended up owing favors to and what those favors entailed etc. You can google Fadil Akgunduz, Mervan Gul with the right keywords (‘pisman’ ‘hapis’ etc.) to piece at least a part of that story together.

  8. With electoral keywords like that, what could possibly go wrong in this country?!

  9. Indeed. Other useful keywords would be irtikap, rusvet etc. Depending on the age of the person, katliam also may produce revealing stuff . Intihal is also very useful about people with academic pasts and is even applicable to the name of a prominent university. (I won’t resist the temptation to point out a certain VP of a certain superpower also deserves the title of muntahil — these people seem to be the same sub-species with superficial adaptations to various geographic regions making identification somewhat difficult).

    All that said, one could observe that despite those — I’d claim indisputably demonstrable– little bits of fun info and what they imply, the country functions reasonably well. It is just that when attempting to analyze the goings on and what’s said, one needs to keep Humpty Dumpty semantics in mind for the jargon used.

  10. […] politics person. Aengus Collins, my favorite ligatured journalist, has some great theoretical stuff written at his blog, which you should be reading […]

  11. […] Why Turkey’s new constitution will be an opportunity missed by istanbulnotes […]

Leave a Reply

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

WordPress.com Logo

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

Google+ photo

You are commenting using your Google+ 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

%d bloggers like this: