Taking land from indigenous people to build a country is historically normal. Countries that built themselves centuries ago now have luxury of criticizing Israel. The US took Mexico only a century and a half ago, and Italy annexed the Vatican decades ago. It is unrealistic to expect that human mentality or historical forces have changed in so short a time. Israel could establish itself similarly at the expense of the locals.

People naturally strive for survival, even if at the expense of others. When chances of individual survival are low, people assemble in groups. Individuals cooperate inside groups (with neighbors) and compete between groups. Groups strive for dominance. They are an instrument of competition, and resort to cooperation only to make a super-group to compete with other large groups or blocs.

Cooperation is peaceful. People don’t need a group in order to cooperate. Specifically, they don’t need a fixed group such as a state, but cooperate in associations which appear ad hoc, fluctuate in terms of membership, and easily dissolve. Fixed groups exist only for hostile competition.
The neighbor-other, individual-group order is not bad. That people don’t fly like birds is normal, not bad. The categories of good and evil apply only to matters of free will.

Groups abstain from competing only when competition is not feasible, the rewards are negligible, or the risks tremendous. States so far are land-based; no virtual states have emerged yet. Even in post-agricultural economies when land lacks huge economic value, it is still indispensable to states as a matter of glory. States naturally seek to enlarge their territory; only the cost of expansion limits their ambitions. Neighboring states, therefore, are always hostile.

Wars start when one state perceives that it has an advantage over another. Peace endures while such an advantage cannot possibly be imagined. Two things impede the advantage: moral backlash or the threat of physical retaliation. The US doesn’t annex, for example, Seychelles, because doing so would run counter to the American raison d’etre of acting as an international gendarme and a beacon of freedom. The Arabs don’t attack Israel simply because they are fearful of retaliation.

The US is a rare moralistic state. America—a supernova—still preserves the moralistic garb of its founders, who were missionaries, conquerors,and settlers. Mature states shed moralism in favor of national interest and expansion of influence.

States refrain from expansion when it brings no significant glory and is likely to provoke concerted opposition. If, hypothetically, Britain embarked now on a spree of annexing island nations, the larger countries would be concerned with the concomitant changes in the worldwide balance of power, and would oppose the British expansion.

Most neighbors don’t annex or otherwise project their influence into each others’ territory only for the fear of retaliation, whether individual (Mexico is afraid to try to take back California) or collective (Germany temporarily ceased fighting for Alsace-Lorraine after it lost to the Allies in WWII). Fear of credible and overwhelming retaliation cements peace.

Perceived changes in the balance of power easily overturn the fear. Throughout history, societies could destroy their neighbors with limited damage to themselves by a certain degree of military preparation and good strategy. The Cold War changed all that. Now a state could assure everyone else that attacking it could not possibly be profitable. Mutually Assured Destruction made even the best conquest unfeasible: a victor could win the war, yet be destroyed. For MAD to work, the retaliation must be credible, unavoidable, and overwhelming.

When a country declares that it won’t use nuclear weapons first, the enemy might believe it. Nuclear deterrent loses its credibility. Enemies begin to imagine that they could attack and not suffer nuclear retaliation. Limitation of liability suggests risky policies.

A country that has only a few small nuclear bombs can’t make the threat unavoidable. But more important than the interception of bombers is the country’s track record of succumbing to international pressure. An attacker could imagine that other countries would persuade the victim to avoid using its nuclear arsenal. That happened with Israel in 1973; a nuclear power risked its very existence, but did not employ nuclear weapons.

Overwhelming retaliation means total destruction of the enemy country. A medium sized 100kt nuclear bomb would kill about 120,000 people if dropped on a city. For a Muslim nation of 60 million people, that’s negligible damage. Even 130 bombs would kill only a fraction of the enemy population, and inflict moderate radiation damage on rural areas. The enemy must be assured in advance that the nuclear retaliation will indeed be overwhelming, completely destroying the attacking army and its depots, as well as all religious and industrial centers.

Retaliation has another dimension: an aggressor should not be able to hide behind a proxy. Soviet-American proxy wars in Africa and the Iranian war against Israel through Hezbollah provide examples. Nuclear retaliation should target the real culprits, and when those culprits are unknown or uncertain, the Islamic enemy collectively should be targeted instead.

The certainty of overwhelming nuclear retaliation prevented the Soviet-American war. MAD is the only approach that will keep Israel safe in a sea of hostile Muslim nations, and that will do so very economically.