Syria has insisted that the question of the Golan Heights be placed on the table at Annapolis, but the future of the basalt highlands half a world away may not be the black-and-white, all-or-nothing rubric that many on both sides present. If Israel acts wisely, it has some tactical wiggle room that can maintain a defensive edge while still offering an opportunity for negotiation. Given the nature of Israel's and Syria's past relations, it will be necessary - no matter what - for Israel to retain the ability to keep Syria from occupying the Heights. To this end, there are a few red lines that Israel must insist upon in any negotiated withdrawal. Some might argue that with modern technology, the Heights under Syrian control are not the threat they once were. But, from the edge of the Golan overlooking the Galilee, Syria's artillery would have a virtual stranglehold on Israel's far North. With the advances in Surface to Air Missile (SAM) technology over the last 40 years, the Israel Air Force would find it more difficult than in the past to silence the Syrian guns or provide support for an assault up the Heights. Therefore, even if Israel were to withdraw from the entirety of the Golan, the Syrian military must not be allowed to occupy it. If the Golan were completely demilitarized, any future war on that front would resemble a race, with both sides sprinting up the Heights and clashing in the middle. In most scenarios stipulating a demilitarized Golan Heights, Syria is the more likely to open a conflict. This would give them a time advantage, which in turn might allow them to place SAMs in the Golan and achieve air neutrality - a situation in which neither side's air force can effectively operate - over the Heights. Most of the key high points in the Golan that would be ideal locations for Syrian artillery (under SAM cover) are on the eastern side of the land mass, making it easier for Syria than for Israel to "grab" them during the initial race to take control of a demilitarized Golan. Those high points, from Mt. Hermon in the North to Tel Fahr in the South, allow their "owners" to cover the rest of the softly descending slope of the Golan and stymie an Israeli counter-offensive. Therefore, in order to guarantee its ability to defend the Golan - and northern Israel - any withdrawal from the Golan by Israel must be partial. But how much can Israel live without? There may be ways in which Israel can withdraw from some areas of the Golan and still retain the ability to deny the Syrian military use of the Heights. The Golan - when considered correctly - can be something close to a defender's dream. It is anchored by difficult terrain and third-party borders. On the North sits the eminently defendable Mount Hermon, and on the South the Golan overlooks the steep decline towards the Yarmuk River. For Syria to outflank either of these positions, it would need to invade either Lebanon or Jordan on the way. Although an invasion of Lebanon as a means of accessing Israel is not entirely out of the realm of possibility, it is unlikely for a myriad of reasons not discussed here. An invasion of Jordan, meanwhile, seems all but out of the question. Therefore, if Israel retained a contiguous defensive line from the Hermon to the Yarmuk, Syria would be forced to assault up the Golan in any future war. In the northern Golan, there are places where Israel could give up a few kilometers of territory - preferably to an extension of the UN-administered demilitarized zone (DMZ) without sacrificing any significant terrain advantage. In the area directly in front of the Hermonit Ridge, for example, Israel could pull the DMZ almost 2.5 kilometers closer and still retain a height advantage of 100 to 200 meters. What is important in the northern Heights is that Israel retain control and maximize its use of the high points that form a rough and jagged line stretching from just north of Hermonit southward through the area of Kibbutz Merom Golan and Tel Fahr. Behind this line, there are no other major terrain features that form as strong a contiguous defensive position. For the most part, the terrain declines until it reaches the pre-1967 border. Between Tel Fahr and the Yarmuk, Israel holds a slight height advantage over the DMZ. But as the line approaches the Yarmuk valley, the DMZ is some 10 meters higher than the Israeli side. In these areas, giving more land to the DMZ might not have any serious impact on Israel's ability to defend the Heights so long as the concessions do not create a deep enough pocket to bypass the southern anchor of Israel's position along the Yarmuk gorge. Whether Israel chooses to withdraw to these lines, remain in its current position, or withdraw even further, its defense capabilities along the Golan Heights would be maximized if it were to invest seriously in heavy fortifications. It is unlikely Syria would be able to bypass the Golan, especially if taking the Golan Heights were the object of the war. It is true that in the Yom Kippur War there were fortifications along the Golan, but it appears that they were primarily designed to give shelter from bombardment rather than to break the impact of a full-scale Syrian assault. They had neither the defensive capabilities nor supplies to withstand a determined Syrian assault. Ideal fortifications would correct both these flaws. The fortifications should be largely underground to withstand the weight of any initial Syrian bombardment and should include heavy artillery - especially on top of the various hills and heights - as well as anti-tank positions. They should be mutually supporting (maintaining visual range to create a web of firepower and communications) and should be based on the "citadel" concept, meaning that each major hilltop position would be able to hold strong even if the lowland positions between it and the next "citadel" were overrun. The positions should be further augmented with mines and other measures to maximize the time in which the defenders could engage their targets at the ideal range. These heavy fortifications would serve to do what the Armored Corps did during the 1973 war in the same area - namely, blunt and stop the Syrian assault. Such fortifications would - even if Israel did withdraw from the areas previously discussed - slightly maximize Israel's terrain advantage and buy the IDF precious time to mobilize for a counter-attack, negating some of the negative consequences of moving the DMZ further into Israeli territory. Moreover, well-built fortifications do not need to be manned by elite infantry. Even if only the northern Golan were heavily fortified, it would free up armored corps and the more elite infantry brigades to concentrate their efforts in the southern Golan. And if the whole Heights were guarded by a series of positions, elite infantry and armored corps forces could concentrate on a counter-offensive. Significant fortification of the Golan has two other advantages. First, it would add a further physical permanence to Israel's continued existence on the Heights - another factor in Israel's favor should future negotiations be considered. Secondly, if the Golan were fortified well enough to guarantee that any Syrian assault would result in extremely high casualties, well-maintained fortifications would serve as a deterrent. It would be foolish to withdraw entirely from a place so strategically important and eminently defendable. If some withdrawal must occur, then it must stop at those points most necessary for the defense of the Heights. Moreover, as long as Israel has the Golan, it should maximize its defensive potential through fortifications in order to assure that it will remain there. Jacob Stoil is the operations manager and research assistant at the International Center for the Study of Radicalization and Political Violence. www.icsr.info.