Tunisia’s draft constitution: Contention over parliamentary vs. (semi-)presidential

A draft, in English, of the Tunisian constitution has been posted at Constitution Net. The draft includes the competing provisions that are being debated by various factions in the Constituent Assembly. Most of the remaining contention concerns the executive format.

The Ennahdha party, which won by far the largest share of seats in the Constituent Assembly election of October 2011, but short of a majority, is campaigning for parliamentarism. This is not surprising, for if the fragmentation of the rest of the party system continues, Ennahdha would be almost assured of holding the prime ministership in a parliamentary system. However, it might struggle to win a presidential election, given that Ennahdha was short of a majority of the vote, and that the opposition might more easily come up with an electable individual candidate than coordinate on an alliance for parliamentary elections.

The two main options are parliamentary and semi-presidential. Under the former proposal, the head of state would be a president selected by the parliament, and all major executive authority would be vested in the prime minister, who along with the cabinet depends on the confidence of the parliamentary majority. Under the latter, the head of state would be popularly elected by two-round majority (see Art. 45).

Moreover, in the proposal with the elected presidency, it would be this president, more than the premier, who would have significant executive power. For instance, the semi-presidential proposal grants the president a veto. If the law is one that the constitution classifies as an organic law, it would take two thirds to override the veto. Otherwise, it would take an absolute majority (half plus one). Even the latter is stronger than the veto that the (unelected) president would have under the parliamentary proposal, which is suspensory only, requiring no larger majority than the original passage of the bill. In case of passing the bill over the objections of the president, it is the president of the chamber, rather than the president, who promulgates the law in the parliamentary proposal (see Art. 57).

Both proposals feature a constructive vote of no confidence (Art. 71), by which the cabinet can be voted out by a majority of legislators only if, on the same vote, the majority elects a new prime minister. Both proposals say that it would take one third of members to initiate a no-confidence vote.

However, the semi-presidential proposal places restrictions on no-confidence (or “censure”) votes :

In the event the specified majority is not attained, the motion of censure may not be reintroduced against the government except after the elapse of a six-month period.

In the event the specified majority is not attained, the motion of censure may not be reintroduced against the government except after the elapse of a six-month period.

I do not see any provision clearly granting the president the right to dismiss a prime minister and cabinet unilaterally. In fact, in addition to the confidence vote provision, both versions at Art. 67 say, “The government shall be held accountable before the Chamber of Deputies.” Without a provision empowering the president to dismiss a cabinet, the system does not meet the criterion of president-parliamentary, the subtype of semi-presidentialism in which the president has a more dominant constitutional role (compared to premier-presidentialism, with its exclusive accountability of the premier and cabinet to the legislative majority). Nonetheless, I might be tempted to classify this proposal as president-parliamentary, given the limits on government responsibility in Art. 71. It would seem that somehow the cabinet must remain responsible to someone during the periods when the parliament can’t engage the government’s responsibility, and that someone would be the president (even if de facto).

The semi-presidential version also allows no-confidence votes against individual ministers, which would seem to undermine the core parliamentary principle of collective responsibility. Such provisions are, however, common in president-parliamentary systems (but not, I think, in premier-presidential ones).

The president, on the other hand, would have little discretion in selecting a prime minister, having to grant the leader of the largest “electoral party or coalition” the first right to form a government after an election (Art. 66). This almost seems, in the present Tunisian party-system context, like a provision meant to maximize the chance for cohabitation! ((Of course, if other parties hold out for a coalition including parties allied with the president, then the resulting cabinet would not be a cohabitation cabinet. But my point is that the constitutional provision appears to give more leverage to the head of the largest party than to the president, which is unusual in a semi-presidential system. An exception that comes to mind is Ukraine, between 2005 and 2012, but in a very different party-system context.))

There is a potentially good argument to be made for a popularly elected counterweight to the dominant Islamist party that has emerged in Tunisia. However, the semi-presidential proposal currently under consideration would seem to risk confrontations over powers between the presidency, on the one hand, and the government and assembly, on the other hand.

11 thoughts on “Tunisia’s draft constitution: Contention over parliamentary vs. (semi-)presidential

  1. “The semi-presidential version also allows no-confidence votes against individual ministers, which would seem to undermine the core parliamentary principle of collective responsibility.”
    What are the consequences of this?


  2. Votes of no confidence against individual ministers are standard, although rare, in Westminster system. They were once much more common, but prime ministerial dominance of the cabinet has made them an expensive luxury for oppositions.


  3. Voting no confidence in a particular Minister would make sense if the legislature had a particular grievance against that official.

    In 1998, the Queensland Parliament (where an independent had the balance of power) voted no confidence in the Attorney-General over an ethics allegation. However, the Premier toughed it out and refused to sack the AG, and the Parliament (or, more accurately, the Independent) wasn’t prepared to force the issue by voting out the entire govt. In our system, though, a NoCoVote in an individual Minister (or even in the whole Cabinet) has no legal force on its own, and is ignored by the Governor if it doesn’t foreshadow the blocking of supply. Other systems slice the goose differently and give NoCoVotes immediate, direct, self-executing legal effect. I tend to think the best balance is to do that for votes against the entire Cabinet but leave votes against individual Ministers a political matter.

    (As for picking off individual portfolio-holders, this seems – ironically – more pronounced in the US system than in Westminster and other parliamentary systems. The departure of Donald Rumsfeld, for a recent example, was attributable to him losing the confidence of the Senate more than to him losing the confidence of the President. At the same time, there were not the numbers in the Senate to deny re-confirmation to other Bush Cabinet-level officials, let alone to try to impeach the President).


  4. … Likewise, the Independent MLAs who held the balance of power in NSW 1991-95 were prepared to vote out Premier Nick Greiner over the Metherell appointment scandal, but had no intention of forcing the entire Liberal/National Coalition out of office, whether by installing the Labor opposition or forcing an early election. (In fact, they had pushed Greiner to agree to introduce semi-fixed terms for NSW).


  5. @#3

    I could not disagree more. A vote of no confidence does not need to be backed by refusing supply.

    If it were the government would be responsible to the parliament only during brief periods when money bills were being debated. Moreover it would make governments responsible to the senate or legislative council, not the lower house.The vast majority of government resignations in Australia have not been tied to refusal of supply and supply has never, to my knowledge, been refused by a lower house in Australia.

    I accept there have been cases like the Gair government resigning on passage of an amendment to a money bill. I think the amendment reduced the premier’s salary by 1 pound.


  6. Alan @5, not sure what you mean. Precedents exist like Canada in 2008 when the Harper Govt lost a vote of confidence but hung on (by proroguing the House) until the numbers changed.

    Legally, what counts (in Westminster-derived monarchical constitutions like the UK, Australia and Canada) is refusal of supply. A motion by the Lower House presaging an eventual blocking of the budget is like a leadership spill by the caucus of the majority party. It gives the Crown/ viceroy a heads-up that the incumbent will probably not be able to get the next budget passed, in which case the Crown/ viceroy may choose to act right away – or alternatively, to place the chief minister on probation.

    PS: I think Curtin vice Menzies in 1941 was also a motion to reduce the PM’s salary by one pound.


  7. This is spinning off the topic of Tunisia, but I’m pretty sure that Tom Round’s Canadian example is incorrect. Harper got the Governor-General to prorogue Parliament to postpone a confidence vote. It wasn’t to get him a re-vote after losing a confidence vote, a request which would in fact have been refused.


  8. Ed’s point @7 taken. Harper lost a vote on an important matter (party funding) that was argued by his opponents to be an implicit matter of confidence, and the warning shot for an explicit vote of no confidence that would have followed had Mme Jean not prorogued. The same cross-party alliance did eventually bring Harper down in 2011.

    So really there are three “levels”:

    (a) defeat on an “important” bill – open to various interpretations – it may be that cross-bench (or floor-crossing Govt) MPs genuinely disliked the proposed law on its merits but don’t want to turf out the PM

    (b) a motion saying “This House lacks confidence in the PM” or “The PM’s salary be reduced by one dollar” or variation thereof – hard to interpret as anything other than “In the name of God, go!”


    (c) refusal to pass a budget bill.

    Tunisia has done better by giving explicit no-confidence votes direct legal force, I’d say.


  9. Any government defeat on a proposed bill is not considered to be a vote of no confidence under the Westminster system, especially if by Westminster you mean the UK. Wikipedia has a list of defeats in the House of Commons suffered by governments since World War II (Wikipedia).

    The Callaghan government, admittedly a minority government, somehow survived for three years despite being defeated a record 34 times.

    In cases where the Constitution codifies what constitutes a “no confidence” vote, its usually done in a way to make it somewhat difficult for the legislature to actually remove the government, at least that is the case with the French and German constitutions.

    Wikipedia also has a helpful list of the governments, in all countries, that were defeated in votes of no confidence:


  10. I know I’m a broken record on this subject but it’s so much easier to just empower the assembly to dismiss the government directly.


  11. Ed @9, I said defeat on a [quote]”important”[unquote] bill.

    Alan @10, I tend to agree but a number of senior Australian and British lawyers seem to think that would lead to dictatorship.


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 )

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.