undocumented references to afErmativenaction — Category 3 —laws on behalfnof its Sephardic minority. Exceptnthat the Sephardics may well be in anmajority, which would rudely shiftnthese actions into Category 2.)nSowell also persistently refuses tonface the difficult questions. Thus, inndiscussing Category 2, “Majority Preferencesnin Minority Economies,” henconfronts situations where the numericalnmajority acts through governmentnto hobble economically dominant minoritiesn(the Chinese in Malaysia, thenLebanese in Sierra Leone). Sowellncriticizes these actions because then”overrepresentation” of, say, Chinesenin top Malaysian economic positions,nwhich is obviously not due to governmentnprivilege, must have been legitimatelynachieved by merit. I agree withnSowell, but it is surely incumbent onnhim to consider the common counterargument:nthat these minority groupsnenjoy the support of their own “clannish”nnetworks, that they find, hire,nand promote each other, and that,nusing such networks, they arrive atntheir socially privileged positions illegitimatelynand at the expense of thenmajority. The point is not that thisnargument is correct, nor that it justifiesnpolitical action against the “clannish,”nonly that a serious scholar must considernand rebut it. But it is never mentioned.nFor an example of deeply flawednanalysis even where considerable worknwas done, let us take Sowell on Ceylon/SrinLanka. Before and after Britainngranted Ceylon independence inn1948, everything, according to Sowell,nwas racially and ethnically harmonious.nBut then, out of the blue, the Sinhalesendemagogue S.W.R.D. Bandaranaikeninjected the irrationality of Sinhalesenracism into Sri Lankan politics inn1956, and the result has been bloodynconflict ever since. If only all partiesnpursued “reason” and “compromise,”naccording to Sowell, they couldnachieve consensus under free-marketnand non-group-preference policies,nand all except a few troublemakersnwould be better off.nIn this we see not only the vagariesnof Thomas Sowell but the deficienciesnof free-market economists in general.nThus, to Sowell, problems of languagenand culture are either bizarre and irrationalnor else simply masks for theneconomic privileges of various ethnicngroups. And yet, as we have seen mostnrecently in the flare-up over Quebecnand its relation to Canada, it makes anbig difference to everyone which languagenis used in courts, schools, andnstreet signs. Most people love theirnlanguage and resent proceedings innsome other tongue. There is nothingnpeculiar or irrational about that. Thengoverning elite of Ceylon had beenneducated in English, in British schools.nA governing elite speaking and writingnEnglish made a great deal of sense sonlong as Ceylon was a colony of Britain;nbut it made no sense whatever oncenCeylon had become independent SrinLanka. Seventy percent of the peoplenof Sri Lanka are Buddhists who speaknSinhalese; 20 percent are Hindus whonspeak Tamil. Only a small fraction ofneither group speak English, and, tonirritate the majority Sinhalese evennfurther, most of the English-speakersnhappen to be Tamil. There could hardlynbe a situation more inherently explosive,nyet Sowell puts the whole problemndown to irrational emotionalism injectedninto politics by Bandaranaike, annirrationality that somehow took on ancontinuing life of its own.nThe point is that independence fornSri Lanka, while an admirablenthing on many grounds, made fiercenethnic conflict inevitable. It is typical ofnSowell’s failure to understand linguisticnand ethnic conflict that he dismisses thenTamil independence movement as thenwork of “extremists.” While it is truenthat Tamil independence — in fact, ethnicnpartition of the island—would notnsolve the problems of the ethnicallynmixed areas in Sri Lanka, it would settlenmuch of the problem by separating thenconcentrated Tamil area in the northnfrom the rest of the island. Ethnicnpartition is no panacea, but it greatlynreduces the area and scope of socialnconflict, which is something that freemarketeersnare supposed to favor. ButnSowell, like most free-market economists,nis a utilitarian democrat, and likenmost utilitarian democrats, he believesnthat everyone’s ends are roughly thensame. In that case, sweet reason canneasily prevail. But in what language isnthat sweet reason to be expressed?nTo put it another way: democraticntheory tells us that the will of thenmajority should prevail. But then, overnnnwhat territorial area should voting expressivenof that will take place? Andnwhy shouldn’t the Tamils of northernnSri Lanka be able to vote all by themselves?nBut if the democrat is to takenthe stern position that no secession ornpartition must ever be tolerated, thennby what logic shouldn’t the territory ofnthe will of the majority be expanded tonthe entire world? In other words,ndoesn’t adamant opposition to any secessionnimply one-world government?nAnd if Thomas Sowell or any otherndemocrat is prepared to go the full waynand advocate world government, thennin what language should its deliberationsntake place?nSowell’s blind spot on the languagenquestion also deeply affects his analysisnof Assam, a large state in northeasternnIndia. Sowell is openly contemptuousnof the affluent though “indolent” Assamesenpeasantry who, eady in Britishnrule, objected strenuously to the competitionnof the immigrating Bengalis.nHungrier and more enterprising, thenBengalis soon grew to economic dominancenin Assam. It is easy to tut-tutnself-righteously over the envy and badnsportsmanship of the Assamese, untilnwe realize that Sowell, once again,npooh-poohs the importance of the languagenproblem. For the Assamese understandablynbridled at Bengali rapidlynbecoming the major language of thenschools and of the ruling elites. Is thisnmerely irrationality and mean-spiritedness?nIs it really so absurd for thenAssamese to wish to continue as anmajority in Assam and to preserve thendominance of the Assamese language?nIn short, mere economic analysis,nlooking at ethnic conflict solely as andesire for special economic privilege bynthe less competent, misses a crucialndimension of the problem. And thisnbrings us straight to the immigrationnquestion. Is it really impermissible for anhome population using its traditionalnlanguage to wish to retain its majoritynstatus in its own homeland? Apparently,nThornas Sowell believes so. Butnthen what would he say to the Latviansn(or Estonians or Moldavians) who havenseen the Russians employ over the pastnhalf-century a deliberate policy of Russificationn— the destruction of the indigenousnlanguage and culture by encouragingnthe immigration of massesnof Russians, who have now reachednsomething like 50 percent of the popu-nDECEMBER 1990/33n