Divided cities are ugly cities: Think of Belfast, of Nicosia, of pre-'89 Berlin and pre-'67 Jerusalem.
There's a reason for this: Cities, unlike other political entities, tend to grow organically, and to sunder one part from another is like sawing a branch off a tree, or a leg off a man. It's not something that, in the natural course of things, ought to happen.
Yet this is what is being proposed as the fence goes up around, and perhaps through, Jerusalem. If Yossi Beilin were to have his way with the Geneva accord, the division would run right through the city's heart. Ehud Olmert has a slightly different idea: He wants the fence to wall off the city's peripheral Arab neighborhoods, like Isawiya (near French Hill), and Umm Tuba (near Ramat Rahel), while maintaining Israeli sovereignty over more central Arab neighborhoods such as the American Colony.
What is to be said about this? First, that it is an ugly idea: ugly in its premises, ugly in its ramifications. And it is one that with better leadership from people like Olmert (who spent nine years as mayor railing against division), might have been avoided. Still, the issue is upon us, and must be debated on the merits.
There are two. One is security: Some of the terrorism that has afflicted the Jewish side of the city came from neighborhoods Olmert wants outside the walls; for example, the killers of French Hill jogger George Khoury were from Kfar Akab. Then there is demography. In 1967, Arabs made up 26 percent of the reunited city's population. In 2000, the percentage was 31%. By 2020, it will be 38%. The Arab birth rate in Jerusalem is three times that of the Jewish birth rate - and Jews are net emigrants from a city where jobs are scarce and housing is expensive.
These are powerful arguments. Yet they are also contradictory. If demography is the great worry, then it makes little sense to propose, as Olmert does, to exclude 50-60,000 Jerusalem Arabs from the city. After all, another 160,000 Jerusalem Arabs would remain on the Israeli side of the fence. More sensible in that case would be to adopt something like the Beilin or Barak plans and split the city right down the middle.
On the other hand, a city split along demographic lines would pose enormous security hazards. Jewish Jerusalemites would be subject to gunfire from inaccessible Palestinian neighborhoods, just as they are in Gilo, and at closer range. And beyond such practicalities is the question of the status of the Old City and the Temple Mount; is this, too, to be relinquished to offload a few thousand more Arabs into Arafat's hands?
In fact, the closer one looks at the case for redividing Jerusalem, the worse it looks. Disengagement - which we favor - is above all an argument about putting Israel behind maximally defensible lines. But just because withdrawal in Gaza is justified does not mean that withdrawal on all fronts is also justified. And this is particularly true in Jerusalem, which must never again be the isolated and besieged enclave it was prior to 1967.
It is a tribute to Israel that as the wall goes up, most Arab Jerusalemites want to be on the Jewish side of it. Israel clearly needs to separate itself from the majority of Arabs. But it neither can, should, nor will, separate itself from all of them. Nowhere is this more true than in Jerusalem, and the great task before Israeli leaders should not be where to divide the city, but how better to bring it together.