Other irrationalities of the current primary process deservernat least brief mention. Crucial contests turn on self-selectedrnprimary electorates, unrepresentative of their states in the generalrnelection and disproportionate in their influence on thernprocess because of their effects on momentum with the nextrncontest. Fifty percent of the voters in the Maryland primaryrnwere at the college-graduate or post-graduate level in education.rnBlue-collar voters simply did not show up in that primary.rnThe Florida primary was dominated by voters over agern60 and by Jewish voters, groups that were both susceptible tornClinton’s negative ads about Tsongas on Social Security andrnon Israel.rnThe social conditions for face-to-facerninteraction and deliberation present on arnsmall scale are not present in the largerrnnation-state. In primaries, referendums,rnopinion polls, and “teledemocracy”rn(such as the “electronic town halls” withrnviewer call-ins advocated by Ross Perot), wernget the isolated, atomized citizen, pulling arnlever, casting a ballot, or dialing an 800 numberrnbased on very little reflection or interaction.rnWhatever the merits of New I lampshire as a starting point,rnwe now expect masses of voters around the country to passrnon candidates about whom they have only the vaguest impressions.rnAs the process approached Super Tuesday, a NewrnYork Times poll of likely primary voters nationwide reportedrn”no opinion” levels for the five leading Democratic candidatesrnranging from 52 to 75 percent. Note these are not levelsrnreflecting that the voters are undecided, but whether they hadrnany opinion at all about these candidates. Furthermore, tornthe extent that voters acquired an opinion, it was likely to bernformed from impressions of smears in supermarket tabloids orrnimpressions that a candidate has the image of being substantive,rnwithout the voters knowing very much about the issues atrnthat stage.rnIowa and New Hampshire are, of course, unrepresentative ofrnthe country—ethnically, racially, politically, on almost anyrndimension you choose. The accident that New Hampshirernis in dire straits economically conditioned the entire beginningrnof this campaign. Imagine if Harkin had not run and if relativelyrnprosperous Iowa had started things off how differentrnthe initial debate and the framing of issues would have been.rnRemember that in 1988, Bush was bounced around by an economicallyrndepressed Iowa but restored by what was then arnprosperous New Hampshire. The arbitrariness of the orderingrnof these events and the conditions in any one state make it imperativernthat we have a different way of starting things off.rnOne study concluded that in 1988 media coverage of Iowarnand New Hampshire voting equaled coverage of all the otherrnprimaries and caucuses put together. Political scientistrnWilliam C. Adams has calculated that New Hampshire typicallyrngets 105 times the coverage per voter as the large Ohiornprimary later in the process.rnAs sociologist Kiku Adatto demonstrated in a much publicizedrnstudy, from 1968 to 1988, the average soundbite forrnpresidential candidates—the period during which a candidaterncould speak uninterrupted on the network evening news—rnshrank from about 42 seconds to a little more than nine seconds.rnThe amount of political discourse reaching the electoraternis being reduced to messages worthy of fortune cookiesrnor bumper stickers. If this election was true to form, all thernminutes of substance on any of the networks’ evening newsrnbroadcasts from January to June will add up to less than twornhours. By contrast, the warm-up to the Super Bowl this yearrnwas two-and-a-half hours.rnWhile some published reports claim that the soundbitesrnincreased in length this vear, such claims depend on includingrnPBS and CNN in calculations for 1992, but not for 1988. Politicalrnscientists Marion Just and Ann Creigler (of the JoanrnShorenstein Barone Center at Harvard) have recently completedrna study of the crucial initial campaign period, fromrnFebruary 1 to 18, and conclude that for the three major networks,rnthe average soundbite is not significantly greater than inrn1988. Furthermore, the average soundbite for Brown wasrndown to six seconds; for Tsongas, Kerrey, and Harkin, aboutrnnine seconds; and for Clinton, about 12 seconds. Clearlv,rnduring the crucial initial period of the campaign season, thernnetworks were doing some initial winnowing of the visibility ofrnthe candidates. The Just/Crcigler study of March shows deteriorationrne’en from these levels, in that Clinton’s soundbitesrnshrank during the climactic period of the primary seasonrnto about six-and-a-half seconds.rnThe mass public has little opportunity or incentive to becomerninformed and little incentive to deliberate on the issues.rnOur democratic mechanisms are really operating asrnwhat political scientists from V. O. Key to Ciovanni Sartorirnhave called an “echo chamber,” bouncing back surface impressionsrnfrom television via polls and primaries. We needrnmore than impression management; we need deliberation onrnthe issues that somehow also represents the entire country.rnThe moe toward mass, direct democracy in the large nation-rnstate derives much of its appeal from an image of directrndemocracy reminiscent of the Athenian Assembly, or ofrnthe New England town meeting. But such an appeal is mistaken.rnThe social conditions for face-to-face interaction andrndeliberation present on a small scale are not present in thernlarger nation-state. In primaries, referendums, opinion polls,rnand “teledemocracy” (such as the “electronic town halls” withrn’iewer call-ins advocated by Ross Perot), we get the isolated, atomizedrncitizen, pulling a lever, casting a ballot, or dialing anrn800 number based on very little reflection or interaction. Hisrnor her vote is just one of millions that will have little effect onrnthe outcome. The citizen has little incentive for informedrndebate or for investment in political knowledge. Even thernmost optimistic account of contemporary voter rationality,rnSamuel Popkin’s recent The Reasoning Voter, characterizesrnvoter thinking in terms of “low information rationality”: thern’oter’s motivations can be generalized as “what have you donernfor me lately?”rnThe old source of deliberation was in the elite processes ofrnthe selection system. The dilemma we are concerned with isrnwhether we have politically equal but relatively incompetentrn24/CHRONICLESrnrnrn