The `Treaty Establishing a Constitution for Europe', which failed after the French and Dutch referenda, measured 482 pages. The actual Treaty, without the myriad declarations and protocols, was down to 209 pages. And what were seen as the most essential parts of the Treaty, the first, institutional, part - which was given no name - and the Charter of Fundamental Rights, measured 29 and 13 pages, respectively.

If you would consider the pragmatic aspects of public debate (the logistics of democracy, if you will), you'd have to recognise that it is easier to carry around and discuss a text of 42 pages than a text of 482 pages.

Not to go deep into the details of the Lisbon Treaty, it is plagued by much the same issues, with the added benefit that it was rendered in a completely unreadable format until the EU released consolidated versions of its amended treaties in early May.

An overwrought legal writing style contributed something to this length, but it is mainly a function of wanting to tinker with two things at once: the institutional set-up of the EU on the one hand, and the policies of the EU on the other hand. It was a major mistake for the EU to want to formulate individual policy objectives as part of something called a `Constitution'. It was also a mistake for the EU to carry on much of the policy changes from its moribund `Constitution' to the Lisbon Treaty. It is harder to sell a Treaty as being about `reform' when it is in fact trying to be many things, and to please many interests.

(Leaving aside, for now, the fact that the masters of narrative in Brussels are not exactly quick studies.)

So the requirement that the text be short rests upon the condition that it is exclusively institutional. At that point, we can also call it a `constitution'. This can be done by leaving the current treaties intact, but letting them be superseded and amended by the constitution. That would also eliminate the need to read long amending texts, as the Lisbon Treaty is -- since the principal text is the authoritative version (The treaties are currently the EU and EC Treaty - under Lisbon the EU and FEU Treaty plus the Charter of Fundamental Rights.)

Doing so implies moving the treaties out of the current half-space of international law in which they uneasily exist, and into a European constitutional space.

By implication, the rules for enacting changes to the treaty should be changed, and a constitution should arrange those rules, as well as the rules for its own amendment.

As a precedent, there is an article in the Lisbon Treaty arranging amendment (48 EU Treaty). However, that article mainly arranges the procedure for drafting amendments, while leaving into place the normal ratification procedure. It also fails to provide for the case where only one or a few Member States do not ratify, simply stating that the matter `shall be referred to the European Council'.

Because the European Union still has intergovernmental character to a large degree, and does not command broad and deep support as an independent political entity, it would be necessary to plot the path away from ratification carefully, so that the Member States do not give up their sovereignty without backing from the people.

Here, a number of checks (for treaty changes) come to mind:

• Arranging for the possibility of secession
• Requiring unanimity in the European Council
• Providing an early possibility for a certain number of national parliaments to cancel the amendment procedure
• Requiring a large majority in the European Parliament
• Arranging for the possibility of `line-item' opt-outs on the most sensitive policies, like defence and taxation
• Providing for a Europe-wide referendum on substantial treaty changes and requiring a large (possibly: double in terms of states and votes) majority in favour
• Making the European Union more clearly subordinate to international law
• Retaining and clarifying the requirement that powers are directly attributed in the Treaties, making the requirement more stringent, and pointing out the possibility for the repatriation of powers
• Retaining, at first, a requirement of national parliamentary ratification for changes to the constitution itself

It would be probable that following such checks and a procedure for drafting a purely institutional constitution would slow down integration in some policy areas over a period of a decade, and perhaps longer, when compared to other alternatives - flexible integration, a last-minute save for the Lisbon Treaty, using the full potential of the Nice Treaty... This would be a positive aspect as many of the people have become sceptical about the speed of integration, and on the long term it will be increasingly important for the European Union to have public support.

(1) For instance, the EU requires more transparency, democracy and popularity, but the power for drafting treaties lies mostly with the least democratic, transparent and popular (but most powerful) institution, the Council. The EU has many supranational aspects but treaty revisions remain intergovernmental. The EU has a low salience for people, due to which there is little discussion and knowledge of it, resulting in low interest, low legitimacy and scare reactions when increases in its power are proposed, concluding in failed attempts to acquire greater salience.