from the extremes of tyranny andnchaos, rationalism and irrationalism —ntwo sides of the same coin. Rather thannexpanding musical possibilities, mostnnew notation is actually inhibiting, andnwhat is being inhibited is music. At thenextreme of rationalism, composernEarle Brown has defended serial musicnon the grounds that it is the techniquen”most rationally compatible with andnrelevant to many methods of analysisnand synthesis employed by mathematicsnand the physical sciences today.”n(He does not explain why art shouldnimitate science.)nAt the same time. Brown admits thatnthe early serialist slogan, “total organization,”nis now embarrassing. To rightnthe balance, he feels, serialist composersnmust add an element of “the uncontrollable.”nThis point of view isnmanifested very clearly in his compositions,nwhere serial and aleatoric passagesnalternate. Nevertheless, the radicalncontradiction between these points ofnview persists; he seeks to “balance”nrather than to integrate them, in thenhope of producing “creative insecurity.”nThus, the contradiction betweennthese two points of view, rather thannproviding a stimulus toward their reintegrationnand the consequent rehumanizationnof music, is prized for itsnown sake. To rationalize this contradictionnwithout resolving it. Brown concludes,n”We can serialize, generalize,nmobilize, do anything now . . .nwithout needing to be right.” For thensophisticated composer, then, anythingnis acceptable as long as it is not universal,nabsolute, or timeless.nTaking this still further, other composersnespouse pure irrationalism.nThey view their primary mission as tondepict disorder, confusion, fear, andnabove all, meaninglessness: a lawlessnuniverse, lacking fixed points of reference.nSuch a viewpoint can be describednas “disintegrationism,” but it isnpart of the myth of progress that has allnbut gutted most of the arts in thisncentury.nRoger Davidson is a composer. Hisnbrother, Nicholas Davidson, is anwriter. Both live in New York City.nMEDIAnThe Uses andnAbuses of PublicnOpinion Pollsnby Jon A. KrosnicknThe Case of Louis Harris andnAssociatesnThe most important principle underiyingndemocracy is that thenmajority should mle. But until relativelynrecently, Americans have been poorlynequipped to communicate their wishesnto elected representatives. The principalnmeans for doing so has always beennelections. But elections occur relativelyninfrequently, and they provide nonmeans for citizens to indicate preciselynwhich of a winning candidate’s standsnthey like and which they dislike. Especiallynpassionate individuals have alwaysnexpressed their views directly to theirnrepresentatives, but these people conveynthe wishes of only a small segment ofnconstituents.nThe development of widespread nationalnpublic opinion polling in then1940’s was therefore a major step forwardnfor the citizens of the UnitednStates. Opinion surveys have made itnpossible for the electorate to expressndetailed demands on policy matters tonelected representatives in Washington.nAnd when national crises occur, we cannswiftly send pointed messages to politiciansnabout what we want done. As anresult, public opinion polls are tremendouslynvaluable assets in the conduct ofncontemporary American democracy.nBut as valuable as public opinionnnnpolls are, they are also easily misinterpreted.nPublic sentiment on most controversialnissues is highly textured andnconflicted. It is therefore difficult tondescribe the American electorate’snviews on issues simply in terms of thenproportions of people who favor andnoppose a broad public policy such asnlegalized abortion. As is true for manynissues, public opinion on abortion variesndepending upon many details, includingnwho the mother is, what the circumstancesnof the pregnancy are, andnwho would pay for and perform it.nBecause public opinion is often sonsensitive to detail, a slight change in thenway a survey question is asked cannsometimes greatly alter people’s answers.nFor example, consider the followingnquestion:nWhich of these do you think isnthe main cause now holdingnback greater prosperity in thisncountry: (a) Business Leaders,n(b) The New Deal, or (c)nLabor?nIn a survey conducted in the earlyn1940’s, “Business Leaders” were selectednas the chief obstruction to prosperity.nHowever, when the third responsenchoice in the question wasnreworded to read “Labor Unions” in ansimultaneous survey of a comparablensample of respondents, answersnchanged significantly. In this case, “LabornUnions” were selected as the mostnimportant obstruction. Apparently, people’snbeliefs about labor in general werendifferent from their beliefs about labornunions in particular.nPollsters have learned an importantnlesson from this example and manynothers like it: whenever the results of ansurvey are reported, the exact wordingnof the questions should be reported asnwell. Even slightly rephrasing a questionncan sometimes seriously mislead readersnabout what respondents were asked.nHowever, despite the fact that paraphrasingna question clearly compromisesnthe honesty and accuracy of a pollnreport, doing so is not uncommon. Thenculprits are sometimes journalists, whonchange wordings to make articles readnmore smoothly. But in one recent case,nthe culprit was a survey firm that conductsnmany important polls: Louis Harrisnand Associates.nDuring September and October ofn1986, Louis Harris and Associates con-nFEBRUARY 1990/47n