George Will's legendary passion for the Democratic Party is perhaps equaled only by former president Ronald Reagan's for such things as communism, activist government and progressive taxation. Which is to say that it does not exist.

Thus when Will expresses fear {"Electoral College's Campus Radical," op-ed, June 3} that the proposal of Skip Roberts' Electoral Fairness Project to change the way presidential electors are selected might hurt the Democratic Party by encouraging Jesse Jackson to run as an independent, one might look elsewhere for the true source of Will's concern.

Perhaps it might lie in the fact that were Roberts's proposal adopted, Horace Busby's Republican southern electoral lock, which seems more often than not to produce a 129-0 electoral vote boost for the GOP presidential candidate, might be reduced to a more reasonable 80-49 advantage. For, under the present system, electors (equal to the number of senators and representatives in each state) are awarded on a winner-take-all basis to the winner of each state's popular vote. Roberts' plan would award an electoral vote to the winner in each congressional district leaving only two electors in each state to be decided by statewide vote.

This in turn might mean that the votes of blacks in East Texas, rural Louisiana, Mississippi, Alabama, Georgia and the Carolina Tidelands and of Latinos in southern Texas might somewhat erode the overwhelming GOP electoral college advantage based on the conservatism of the region as a whole.

But the virtues of Roberts' idea lie not in its partisan benefits or dangers but in its potential benefits to the American political system as a whole. For it would address a number of ills including:

Media. Now between 80 and 90 percent of a campaign budget is spent on political advertising and fund-raising for it, leaving next to nothing for activities involving citizens. Under the proposed change, there would be an incentive to devote campaign resources to grass-roots activity in congressional districts in order to capture shares of the electoral vote.

Polls. Now when campaign polls show a candidate substantially behind another in a particular state, the tendency is for both campaigns to move all their resources to more competitive states. Under the proposal, there would be an incentive not to abandon those states and leave some resources and reasons to mobilize the citizenry.

Pluralism. Under the plan, it would not simply be southern blacks and Texas Latinos, whose votes are subsumed under a conservative tide, who might find some incentive to vote. But rural eastern Kentuckians might find their voice heard rather than buried under the votes of Louisville, Lexington and Newport. Upstate New Yorkers may have their more moderate views heard rather than overcome by more liberal New York City. Minnesota farm and Iron Range voters might be able to distinguish their views from the Twin Cities' urban interests. And it would force campaigns to address and accommodate those interests and to mold a true pluralism based on something other than demagogic media appeals.

Parties. The proposal would likely serve to strengthen the two-party system. It would, by enhancing the pluralistic grass roots and deemphasizing polls and paid media, diminish the role of unaccountable political consultants. And it would create the possibility of more coherent party structures and programs.

Party program development would not have to bend so far to accommodate state and regional interests as to render them meaningless. And, by focusing on districts rather than on states, it would permit candidates for local and federal offices to support the programs and policies of their party's standard-bearer rather than running from them as many currently do.

The proposal would also have the minor advantage of making the result of the electoral college vote more nearly mirror the result of the popular vote. And it would make television network declarations of election winners while people are still voting to determine those winners infinitely more costly and complex.

And all this could be accomplished without undermining a major benefit of the present system -- enhancing federalism -- for the 100 votes that could still be won on a statewide basis are too large a prize to be ignored by any presidential campaign.

About 30 years ago, according to Will's colleague David Broder, conservative Republican Sen. Karl Mundt suggested precisely the same revision in the way electors are selected as the potential salvation for the Republican Party. Now Democrats see it as a way to break the southern GOP electoral college lock. Thirty years from now who knows who the apparent partisan beneficiary will be?

But any proposal that addresses many of the serious ills of the modern campaign -- the dominance of polls, consultants and scurrilous media, the erosion of political parties and grass-roots activity, the lack of coherence in our national political dialogue and the declining will of citizens to get involved -- is worthy of serious consideration.

The writer is director of the nonpartisan Committee for the Study of the American Electorate.