OXFORD,ENGLAND — Oxford, England. Old wars like old generals never die, they just fade away. It's only a few years since Ronald Reagan's defense secretary, Caspar Weinberger, claimed in justification of American military involvement in Central America that the region was ''on the mainland of the U.S.'' and that ''defending the mainland ranks above all other priorities.''
This is how it was in the bad old days of the Cold War. We may smile wryly now that we are so absorbed with the Persian Gulf. It is not just disconcerting but disorienting how quickly what is vital can become almost trivial. It is tempting to think that Washington's perennial preoccupation with Central America exists like a permanent subterranean fissure erupting to the surface only when the politicos need something hot and there's nothing grander in the world to absorb their attention.
Right now one can be pretty sure that even if the Sandinistas staged a comeback coup or the Noriega supporters had done even better than they did in last month's local elections in Panama Washington would have found a way to hype it down rather than up.
Doubtless, however, the United States will return to Central America one day because as long as the area remains a seething caldron of social, economic and political instability there will always be some cataclysm or other that serves as an excuse for poking in the military finger. The alternative of systematic long-term aid in proportions sufficient to deal with the social inequities that fuel the culture of disequilibrium never has had the same drawing power in Washington.
Long before Marxist theoreticians gained a following in Central America, the U.S. was prone to send in the Marines. And one can imagine all sorts of post-Marxist scenarios that lead to future intervention: the deterioration of the Panama Canal, a process that is already well under way as American engineers and supervisors are being replaced by less experienced local personnel; or the resurgence of the narcotics trans-shipment as the U.S. moves both coast guards and electronic monitoring equipment to the Persian Gulf theater; or new guerrilla insurgencies. Compare the way the old-time Marxists in El Salvador's Farabundo Marti National Liberation Front fight with the undisciplined Shining Path guerrillas of Peru, who owe allegiance to no one. Latin America has never seen such malicious ferocity since Pizarro subjugated the Incas.
It is tempting to try to rebuff Wahington's evergreen interest in Central American politics with the kind of deflating history Professor Samuel Stone, of the University of Nebraska, has just produced in his new book, ''The Heritage of the Conquistadors.'' His beguiling thesis is that the descendants of the Spanish conquerors are so entrenched in high positions in these feudal societies that there will always be a violent cleavage between top dog, mainly white, and underdog, mainly Indian, and that to push for social evolution, human rights, democracy and stability is but the daydreaming of Western liberals.
He has researched the genealogical tables to prove how closed the world of power politics is in Central America. All but one of Costa Rica's 38 presidents have shared descent from a single conquistador, Cristobal de Alfaro, who was also a progenitor of seven presidents in Nicaragua and one each of Guatemala and Honduras.
In Nicaragua one can see the effects of de Alfaro's dynasty in malignant clarity. The Sandinista rebellion was aimed at the Somoza family that ruled from 1933 to 1980. But who came to power after the Sandinistas were finally sidelined by an unexpected defeat in an election last year? None other than those Somoza cousins, the Chamorros and Lacayos. Did the Sandinistas do more than briefly interrupt a family row?
Professor Stone makes his point well, but it is doubtful that genealogical research on its own will stay Washington's hand in the future or even push it to wind up its commitment to El Salvador's messy 11-year war.
For that, a Pauline conversion is needed in Washington to the viewpoint that while the political arrangements of Central America are of utmost concern to the inhabitants themselves, they are only of marginal importance to the essential interests of the United States. And even where at first sight they appear to impinge on American security -- the Canal, the drug trade, the boundary disputes -- there are always better alternatives to military intervention -- transport blimps instead of the Canal, decriminalization of the drug industry and U.N. policing for political conflict.
It would be nice to think that during the present lull in U.S. operations in Central America, those who are still paid to think about the isthmus might use this time to reflect how little America has gained from militaristic meddling in its own backyard.